CN116883067B - Medical brand popularization system and method - Google Patents
Medical brand popularization system and method Download PDFInfo
- Publication number
- CN116883067B CN116883067B CN202310994205.5A CN202310994205A CN116883067B CN 116883067 B CN116883067 B CN 116883067B CN 202310994205 A CN202310994205 A CN 202310994205A CN 116883067 B CN116883067 B CN 116883067B
- Authority
- CN
- China
- Prior art keywords
- brand
- patient
- medical staff
- medical
- scoring
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 53
- 239000003814 drug Substances 0.000 claims abstract description 82
- 230000005540 biological transmission Effects 0.000 claims abstract description 66
- 230000008569 process Effects 0.000 claims abstract description 23
- 230000000694 effects Effects 0.000 claims abstract description 22
- 238000010276 construction Methods 0.000 claims abstract description 7
- 230000006870 function Effects 0.000 claims description 53
- 230000009471 action Effects 0.000 claims description 31
- 238000013507 mapping Methods 0.000 claims description 31
- 238000013528 artificial neural network Methods 0.000 claims description 25
- 238000012216 screening Methods 0.000 claims description 17
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 claims description 9
- 201000010099 disease Diseases 0.000 claims description 6
- 108010003272 Hyaluronate lyase Proteins 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 3
- 230000001902 propagating effect Effects 0.000 claims description 3
- 230000007480 spreading Effects 0.000 abstract description 6
- 238000003892 spreading Methods 0.000 abstract description 6
- 230000002787 reinforcement Effects 0.000 description 12
- 239000003795 chemical substances by application Substances 0.000 description 7
- 230000006399 behavior Effects 0.000 description 6
- 238000010801 machine learning Methods 0.000 description 6
- 230000009466 transformation Effects 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 229940079593 drug Drugs 0.000 description 3
- 230000006978 adaptation Effects 0.000 description 2
- 238000007405 data analysis Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000001737 promoting effect Effects 0.000 description 2
- 208000024891 symptom Diseases 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 229940127554 medical product Drugs 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- 238000000844 transformation Methods 0.000 description 1
- 230000001131 transforming effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0202—Market predictions or forecasting for commercial activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0277—Online advertisement
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Strategic Management (AREA)
- Finance (AREA)
- Development Economics (AREA)
- Accounting & Taxation (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Entrepreneurship & Innovation (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Economics (AREA)
- Game Theory and Decision Science (AREA)
- Biophysics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Evolutionary Computation (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Molecular Biology (AREA)
- General Health & Medical Sciences (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The invention discloses a medicine brand promotion system and a method, which relate to the technical field of data promotion, wherein the system comprises: the medical brand data input part, the medical staff brand scoring part, the patient brand scoring part, the brand popularization and transmission network construction unit and the brand popularization unit; the medicine matching data input part is configured to input medicine brand data; the medical brand data specifically includes: a pharmaceutical brand name and a brand ID that uniquely identifies the pharmaceutical brand name; and a medical staff brand scoring section configured to provide the medical staff with a score for entering the medical brands, each medical staff, when scored, entering the medical staff age, the medical staff geographic location, and the medical staff duty. According to the invention, the information spreading process is simulated, the optimal popularization strategy is automatically found, the refinement degree and the personalized effect of the medicine popularization can be effectively improved, and the medicine popularization efficiency and effect are higher.
Description
Technical Field
The invention relates to the technical field of data promotion, in particular to a medicine brand promotion system and method.
Background
With the development of society and the advancement of technology, the medical industry is increasingly important in global economy. How to effectively popularize medical products to potential patients and medical professionals is a significant challenge facing pharmaceutical companies and related enterprises. Traditional pharmaceutical promotion is mainly dependent on advertising, sales person face-to-face promotion and other marketing means. However, with the development of new technologies such as big data and artificial intelligence, the way of popularizing pharmaceuticals has also changed deeply.
In recent years, some researchers and businesses have begun to attempt to improve the popularization of pharmaceuticals using data analysis and machine learning techniques. For example, by collecting and analyzing patient health information, consumption behavior data, and data on a social network, the patient's needs can be predicted more accurately, thereby customizing more efficient marketing strategies. In addition, some enterprises have attempted to automatically optimize strategies for pharmaceutical popularization using machine learning algorithms, such as deep learning and reinforcement learning.
However, some problems still remain with the prior art. First, while data analysis may provide valuable insights, it remains a challenge to translate these insights into actual marketing actions. Second, existing machine learning methods often require large amounts of data to be effective, which can be difficult to obtain in the pharmaceutical industry, as the acquisition of medical data is often subject to stringent regulatory restrictions. Furthermore, these methods often assume that the data is static, ignoring changes in medical needs and behavior patterns over time.
In addition, existing pharmaceutical popularization strategies often ignore network effects in the medical ecosystem. In reality, medical professionals, patients, pharmaceutical companies, and other related parties are interrelated and their behaviors can affect each other. For example, a doctor's recommendation may affect a patient's drug selection, while a pharmaceutical company's marketing strategy may affect a doctor's recommendation behavior. Therefore, an effective pharmaceutical popularization strategy needs to take such network effects into account.
In summary, the existing methods for promoting pharmaceuticals have certain problems in terms of data acquisition, processing and utilization, and understanding and utilizing the network effects of the medical ecosystem. Therefore, a new method is urgently needed that can more effectively utilize data and new technologies to solve these problems in the promotion of pharmaceuticals.
Disclosure of Invention
The invention aims to provide a medicine brand popularization system and method, which can effectively improve the refinement degree and personalized effect of medicine popularization and simultaneously have higher popularization efficiency and effect by simulating an information transmission process and automatically searching an optimal popularization strategy.
In order to solve the technical problems, the invention provides a medicine brand promotion system and a method, comprising the following steps:
A pharmaceutical branding system, the system comprising: the medical brand data input part, the medical staff brand scoring part, the patient brand scoring part, the brand popularization and transmission network construction unit and the brand popularization unit; the medicine matching data input part is configured to input medicine brand data; the medical brand data specifically includes: a pharmaceutical brand name and a brand ID that uniquely identifies the pharmaceutical brand name; a medical staff brand scoring component configured to provide medical staff with a score for entering a medical brand, each medical staff, when scored, entering a medical staff age, a medical staff geographic location, and a medical staff job; a patient brand scoring component configured to provide patients with a score for the entered medical brands, each patient, when scored, entering a patient age, a patient geographic location, and an affected condition; the brand promotion and propagation network construction unit is configured to analyze the result of the patient scoring of the recorded medicine brand data and the result of the medical staff scoring at first, and specifically comprises the following steps: screening out medicine brand data of which the patient scoring and the medical staff scoring exceed a set first threshold, screening out medicine brand data of which the patient scoring and the medical staff scoring are lower than a set second threshold, screening out medicine brand data of which the patient scoring exceeds the set first threshold and the medical staff scoring is lower than the set second threshold, or reserving medicine brand data of which the patient scoring exceeds the first threshold as preferable medicine brand data, and reserving medicine brand data of which the patient scoring and the medical staff scoring are in a section formed by the upper limit of the first threshold and the lower limit of the second threshold as less preferable medicine brand data; mapping each patient to a point in a three-dimensional space coordinate system based on the patient age, the patient geographic position and the affected condition corresponding to each patient, mapping each medical staff to a point in the three-dimensional space coordinate system based on the medical staff age, the medical staff geographic position and the medical staff duty of each medical staff, and connecting all points in the three-dimensional space coordinate system as a propagation network; the brand promotion unit is configured to take the preferred medical brand data as a transmission factor, randomly select an initial node in the transmission network, enable the transmission factor to be transmitted in the transmission network under the drive of a set transmission model, enable the number of nodes through which the transmission factor passes to be maximum in a set time interval, record the transmission path of the transmission factor as a brand promotion policy, carry out brand promotion, and repeatedly execute the above-mentioned process on the suboptimal medical brand data after the preferred medical brand data are processed, and carry out brand promotion.
Further, the process of mapping each patient to a point in the three-dimensional space coordinate system based on the patient age, the patient geographic location and the affected condition corresponding to each patient specifically includes: taking the age of the patient as an X-axis coordinate, and taking the mean value of longitude and latitude of the geographic position of the patient as a Y-axis coordinate; assigning a uniquely identified condition value to the affected condition as a Z-axis coordinate; the disease values corresponding to different diseases are different.
Further, the process of mapping each medical staff member to a point in the three-dimensional space coordinate system based on the medical staff age, the medical staff geographic location and the medical staff duty of each medical staff member specifically comprises: taking the age of medical staff as an X-axis coordinate, and taking the mean value of longitude and latitude of the geographical position of the medical staff as a Y-axis coordinate; assigning the medical staff with a uniquely identified duty value as a Z-axis coordinate; the corresponding duty values of different medical staff are different.
Further, the brand promotion unit uses the optimized medicine brand data as a transmission factor, randomly selects an initial node in a transmission network, and makes the transmission factor transmit in the transmission network under the driving of a set transmission model, so that the method for maximizing the number of nodes through which the transmission factor passes in a set time interval comprises:
Step S1: defining a state representation: mapping each medical staff member and patient to a point in a three-dimensional spatial coordinate system, the state s being expressed as (x, y, z), wherein x represents age, y represents geographic location, z represents job for medical staff member, and z represents disorder for patient;
Step S2: defining an action space: randomly selecting an initial node in a transmission network, and taking the preferred medicine brand data as a transmission factor M; and propagating under the driving of the set propagation model; defining an action a as a popularization value when a propagation factor M reaches a certain node in a propagation network; defining a set time interval as T; the promotion value is defined as the number of other nodes in the covered circumference area multiplied by the node weight by taking the node as the center and taking the set threshold value as the radius; when the node is a node obtained by mapping of medical staff, the node weight is A; when the node is a node obtained by mapping a patient, the node weight is B;
step S3: defining a reward function r (s, α, M): the rewarding function r (s, a, M) measures the effect index of the propagation factor M after the state s executes the action a;
Step S4: defining a Q value function: the Q-value function Q (s, a, M) represents the expected jackpot that the propagation factor M would have obtained after the state s performed action a; estimating a Q-value function using a neural network, denoted Q (s, α, M; θ), where θ is a parameter of the neural network;
Step S5: forming an experience playback buffer in a state-action-rewards sequence collected within a set time T; then, updating a parameter θ of the neural network using a DQN algorithm to minimize a mean square error between the Q-value function and the target Q-value function; at this time, the number of nodes through which the propagation factor passes is the largest.
Further, the target Q value function is expressed using the following formula:
Where s' represents the next state after action a is performed, θ - represents a parameter of the target neural network, and γ is a discount factor for balancing the importance of the instant and future rewards; q target (s, a, M) is a target Q function; a' is the next action; p propagation network complexity.
Further, in the step S5, when the DQN algorithm is used to update the parameter θ of the neural network to minimize the mean square error between the Q-value function and the target Q-value function, the mean square error is expressed by using the following formula:
Wherein, For mean square error, s i,ai,r(si,ai, M) represents the state-action-reward sequence for a set time T in the empirical playback buffer, N being the size of the buffer; i is a subscript sequence number, the value is a positive integer, and the initial value is 1.
Further, the update formula of the Q value function Q (s, a, M) is:
where α is the learning rate, used to control the updated step size.
Further, when the node is a node obtained by mapping of medical staff, the node weight A is equal to 0.6; when the node is a node obtained by mapping a patient, the node weight B is 0.4; the spreading factor M is represented using a brand ID of each pharmaceutical brand data.
Further, the calculation formula of the complexity of the propagation network is as follows:
p＝L*(|lg(Pa)|-1)*exp(Do)；
wherein L is the total node number of the propagation network; pa is the number of nodes mapped by the patient; do is the number of nodes mapped by medical personnel.
A method of medical promotion, the method comprising:
Inputting medicine brand data; the medical brand data specifically includes: a pharmaceutical brand name and a brand ID that uniquely identifies the pharmaceutical brand name; marking the brands of the recorded medicines by medical staff, and recording the ages, geographical positions and job positions of the medical staff by each medical staff when marking; the patients score the entered medicine brands, and each patient enters the age of the patient, the geographical position of the patient and the affected symptoms when scoring; firstly, analyzing the scoring result of patients and the scoring result of medical staff of the entered medicine brand data, wherein the method specifically comprises the following steps: screening out medicine brand data of which the patient scoring and the medical staff scoring exceed a set first threshold, screening out medicine brand data of which the patient scoring and the medical staff scoring are lower than a set second threshold, screening out medicine brand data of which the patient scoring exceeds the set first threshold and the medical staff scoring is lower than the set second threshold, or reserving medicine brand data of which the patient scoring exceeds the first threshold as preferable medicine brand data, and reserving medicine brand data of which the patient scoring and the medical staff scoring are in a section formed by the upper limit of the first threshold and the lower limit of the second threshold as less preferable medicine brand data; mapping each patient to a point in a three-dimensional space coordinate system based on the patient age, the patient geographic position and the affected condition corresponding to each patient, mapping each medical staff to a point in the three-dimensional space coordinate system based on the medical staff age, the medical staff geographic position and the medical staff duty of each medical staff, and connecting all points in the three-dimensional space coordinate system as a propagation network; firstly, taking preferable medicine brand data as a transmission factor, randomly selecting an initial node in a transmission network, enabling the transmission factor to be transmitted in the transmission network under the drive of a set transmission model, enabling the number of nodes through which the transmission factor passes to be maximum in a set time interval, recording a transmission path of the transmission factor, taking the transmission path as a brand popularization strategy, carrying out brand popularization, and repeatedly executing the process on less preferable medicine brand data after the preferable medicine brand data are processed, and carrying out brand popularization.
The medical brand popularization system and method have the following beneficial effects:
Firstly, the invention converts the medical brand promotion problem into the reinforcement learning problem by defining the state representation, the action space, the rewarding function and the Q value function. This transformation allows us to automatically find the optimal popularization strategy using reinforcement learning algorithms without the need to manually set complex rules or make extensive manual tuning. The difficulty and the workload of popularization strategy design are greatly reduced, and the popularization strategy can be better adapted to the change of market environment.
Secondly, the invention utilizes the neural network to estimate the Q value function, and can effectively process a high-dimension and continuous state space and action space. This means that our promotion model can handle a variety of complex factors including age, geographic location, job title, condition, etc., thus making the promotion strategy finer and more personalized. Meanwhile, parameters of the neural network are updated through experience playback and an DQN algorithm, so that the problems of sample efficiency and non-stability in reinforcement learning can be effectively solved.
Again, the present invention takes into account network effects in pharmaceutical popularization. The spreading of the spreading factors in the network simulates the spreading process of the medicinal information in the medical ecological system, and the setting of the weight of the mapping nodes of the medical staff and the patients reflects the importance and influence of the medical staff and the patients in the network. The consideration of the network effect enables the popularization strategy to be more in line with the actual situation of the medical ecological system.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are required to be used in the embodiments or the description of the prior art will be briefly described below, and it is obvious that the drawings in the following description are only embodiments of the present invention, and that other drawings can be obtained according to the provided drawings without inventive effort for a person skilled in the art.
Fig. 1 is a schematic system structure diagram of a pharmaceutical brand promotion system according to an embodiment of the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments of the present invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Example 1: a pharmaceutical branding system, the system comprising: the medical brand data input part, the medical staff brand scoring part, the patient brand scoring part, the brand popularization and transmission network construction unit and the brand popularization unit; the medicine matching data input part is configured to input medicine brand data; the medical brand data specifically includes: a pharmaceutical brand name and a brand ID that uniquely identifies the pharmaceutical brand name; a medical staff brand scoring component configured to provide medical staff with a score for entering a medical brand, each medical staff, when scored, entering a medical staff age, a medical staff geographic location, and a medical staff job; a patient brand scoring component configured to provide patients with a score for the entered medical brands, each patient, when scored, entering a patient age, a patient geographic location, and an affected condition; the brand promotion and propagation network construction unit is configured to analyze the result of the patient scoring of the recorded medicine brand data and the result of the medical staff scoring at first, and specifically comprises the following steps: screening out medicine brand data of which the patient scoring and the medical staff scoring exceed a set first threshold, screening out medicine brand data of which the patient scoring and the medical staff scoring are lower than a set second threshold, screening out medicine brand data of which the patient scoring exceeds the set first threshold and the medical staff scoring is lower than the set second threshold, or reserving medicine brand data of which the patient scoring exceeds the first threshold as preferable medicine brand data, and reserving medicine brand data of which the patient scoring and the medical staff scoring are in a section formed by the upper limit of the first threshold and the lower limit of the second threshold as less preferable medicine brand data; mapping each patient to a point in a three-dimensional space coordinate system based on the patient age, the patient geographic position and the affected condition corresponding to each patient, mapping each medical staff to a point in the three-dimensional space coordinate system based on the medical staff age, the medical staff geographic position and the medical staff duty of each medical staff, and connecting all points in the three-dimensional space coordinate system as a propagation network; the brand promotion unit is configured to take the preferred medical brand data as a transmission factor, randomly select an initial node in the transmission network, enable the transmission factor to be transmitted in the transmission network under the drive of a set transmission model, enable the number of nodes through which the transmission factor passes to be maximum in a set time interval, record the transmission path of the transmission factor as a brand promotion policy, carry out brand promotion, and repeatedly execute the above-mentioned process on the suboptimal medical brand data after the preferred medical brand data are processed, and carry out brand promotion.
In particular, for a medical brand, if a medical brand is highly rated by patients and healthcare workers, this means that the medical brand does not already need to be promoted. Therefore, the invention judges and screens by setting two thresholds. If a medical brand is rated less in both the patient and the healthcare worker, the brand is also of no promotional value. Through threshold value screening, medical brands with popularization value can be well screened out. Promote the effect of popularization.
The system starts by randomly selecting a node in the propagation network, allowing the preferred pharmaceutical brand data to propagate in the network as a propagation factor. And recording the paths with the maximum number of nodes passed by the propagation factors in the set time interval as a brand promotion strategy. Based on social network theory, the method considers the information transmission in the social network as a complex process, and can find the optimal popularization path by simulating the process.
The promotion of the medicine brands is regarded as an information spreading process in a spreading network, and data are managed and processed by utilizing the theory and method of data science, so that the promotion activity is more accurate and effective. In addition, the influence of extreme data on the evaluation result can be avoided by using the threshold value to carry out data screening, and the fairness and the accuracy of the evaluation are improved.
Example 2: on the basis of the above embodiment, the process of mapping each patient to a point in the three-dimensional space coordinate system based on the patient age, the patient geographic location and the suffering condition corresponding to each patient specifically includes: taking the age of the patient as an X-axis coordinate, and taking the mean value of longitude and latitude of the geographic position of the patient as a Y-axis coordinate; assigning a uniquely identified condition value to the affected condition as a Z-axis coordinate; the disease values corresponding to different diseases are different.
Example 3: on the basis of the above embodiment, the process of mapping each medical staff member to a point in the three-dimensional space coordinate system based on the medical staff age, the medical staff geographic location and the medical staff duty of each medical staff member specifically includes: taking the age of medical staff as an X-axis coordinate, and taking the mean value of longitude and latitude of the geographical position of the medical staff as a Y-axis coordinate; assigning the medical staff with a uniquely identified duty value as a Z-axis coordinate; the corresponding duty values of different medical staff are different.
In particular, this method can be regarded as a process of feature engineering. Feature engineering is the process of transforming raw data into features that more reflect its underlying structure, often used to improve the performance of machine learning algorithms. In this case, the age, geographical location and condition or role of the patient and medical personnel are extracted as representative of their location in three-dimensional space. This allows for better capture of their primary features and facilitates subsequent analysis and processing.
Connecting these points forms a complex network. This network can be used to simulate and understand the pattern of spread of pharmaceutical brands in society. Therefore, the method can help to find the most effective popularization strategy and improve the awareness and acceptance of the medicine brands.
Example 4: based on the above embodiment, the brand promotion unit uses the preferred pharmaceutical brand data as a propagation factor, randomly selects an initial node in the propagation network, and propagates the propagation factor in the propagation network under the driving of a set propagation model, so that the method for maximizing the number of nodes through which the propagation factor passes in a set time interval includes:
Step S1: defining a state representation: mapping each medical staff member and patient to a point in a three-dimensional spatial coordinate system, the state s being expressed as (x, y, z), wherein x represents age, y represents geographic location, z represents job for medical staff member, and z represents disorder for patient;
Step S2: defining an action space: randomly selecting an initial node in a transmission network, and taking the preferred medicine brand data as a transmission factor M; and propagating under the driving of the set propagation model; defining an action a as a popularization value when a propagation factor M reaches a certain node in a propagation network; defining a set time interval as T; the promotion value is defined as the number of other nodes in the covered circumference area multiplied by the node weight by taking the node as the center and taking the set threshold value as the radius; when the node is a node obtained by mapping of medical staff, the node weight is A; when the node is a node obtained by mapping a patient, the node weight is B;
step S3: defining a reward function r (s, α, M): the rewarding function r (s, a, M) measures the effect index of the propagation factor M after the state s executes the action a;
Step S4: defining a Q value function: the Q-value function Q (s, a, M) represents the expected jackpot that the propagation factor M would have obtained after the state s performed action a; estimating a Q-value function using a neural network, denoted Q (s, α, M; θ), where θ is a parameter of the neural network;
Step S5: forming an experience playback buffer in a state-action-rewards sequence collected within a set time T; then, updating a parameter θ of the neural network using a DQN algorithm to minimize a mean square error between the Q-value function and the target Q-value function; at this time, the number of nodes through which the propagation factor passes is the largest.
Specifically, the brand promotion problem is converted into a reinforcement learning problem. Reinforcement learning is one type of machine learning, with the goal of letting agents learn optimal behavior strategies through interactions with the environment. In this problem, the agent is a brand of medicine, the environment is a network composed of medical staff and patients, the state is the position of the agent in the environment, the action is the promotion activity that the agent can perform at each position, and the reward is the effect brought by each action. Through learning, the agent can find a strategy, so that the number of nodes through which the propagation factors pass is maximized within a set time interval, and the brand popularization effect is maximized.
Reinforcement learning is an important branch of machine learning in which an agent (here a pharmaceutical branding system) learns and optimizes its behavior strategies by interacting with the environment (here a network of medical staff and patients).
Step S1 uses information about the age, geographical location, and job or condition of medical personnel and patients to map them onto points in three-dimensional space, defining a state S. The state representation method is helpful for mining individual characteristics of medical staff and patients and interrelationships among the medical staff and the patients, and provides possibility for accurate popularization of medicine brands.
Action a in step S2 is defined as the popularization value when the propagation factor M reaches a certain node in the propagation network. This arrangement essentially treats the promotion process of the pharmaceutical brands as a decision process: in each state, the system needs to decide what promotion strategy (i.e., action) to take in order to maximize its promotion effect (i.e., rewards).
Step S3 winning function r (S, a, M) measures the effect of spreading factor M after action a is performed in state S. This effect may include, but is not limited to, the breadth of the propagation range, the depth of the influence, etc. The reward function is a key component in reinforcement learning, and by optimizing the reward function, the system can learn to adopt a more effective popularization strategy.
The Q function Q (S, a, M) in step S4 represents the expected jackpot that the propagation factor M would achieve after performing action a in state S. The Q function plays a key role in reinforcement learning, which can help the system evaluate and select the optimal action. Here, the Q-value function is estimated by a neural network, so that the system can be better adapted to a complex real environment.
The system first collects data over a period of time and then optimizes the parameters of the neural network by applying a Deep Q Network (DQN) algorithm to minimize the mean square error between the Q function and the target Q function in step S5. The process is based on two reinforcement learning technologies of experience playback and Q learning, and aims to realize a more effective learning process and improve the popularization effect of the system.
Example 5: on the basis of the above embodiment, the target Q value function is expressed using the following formula:
Where s' represents the next state after action a is performed, θ - represents a parameter of the target neural network, and γ is a discount factor for balancing the importance of the instant and future rewards; q target (s, a, M) is a target Q function; a' is the next action; p propagation network complexity.
This formula is based on the bellman equation in reinforcement learning, which describes the prize value (i.e., Q value) that an agent can expect to obtain after performing action a in state s. Where r (s, a, M) is the current immediate prize, γ is the discount factor, for balancing the current and future prizes, max a'Q(s',a',M;θ-) is the maximum Q value at the future state s'. Here exp (p) represents the complexity of the propagation network, the higher the complexity, the greater the impact on the current prize r (s, a, M).
In particular, in the environment of pharmaceutical branding, the complexity of the propagation network may have an impact on the promotion effect. For example, if the network is very complex, the propagation of information may be more difficult. Thus, by introducing this complexity factor, the model can be better adapted to complex propagation environments.
From this formula, it can be seen that in determining the action, the system considers not only the current rewards and future expected rewards, but also the complexity of the propagation network. This allows the system to better address the particularities of pharmaceutical branding, for example, different medical personnel and patients may have different ways and efficiencies of receiving information, as embodied by the network complexity factor p.
In addition, the parameter θ - of the target neural network is also used in this formula, which is a technique commonly found in deep Q learning (DQN), and is called "fixed qtarget". In order to increase stability during training, a fixed, late updated target network is used to calculate the target Q.
Example 6: on the basis of the above embodiment, when the parameter θ of the neural network is updated using the DQN algorithm in the step S5 to minimize the mean square error between the Q-value function and the target Q-value function, the mean square error is expressed using the following formula:
Wherein, For mean square error, s i,ai,r(si,ai, M) represents the state-action-reward sequence for a set time T in the empirical playback buffer, N being the size of the buffer; i is a subscript sequence number, the value is a positive integer, and the initial value is 1.
Specifically, definition: first, s i,ai,r(si,ai, M) represents a state-action-rewards sequence sampled from an empirical playback buffer, which data was collected over a past time step T. The size of the buffer is N, and i is the index number representing the data.
This formula is a loss function in neural network training, which is used in the DQN algorithm in reinforcement learning in the present invention. In this formula, an attempt is made to minimize the error between the predicted Q value Q (s i,ai, M; θ) and the target Q value Q target(si,ai, M), multiplied by the corresponding prize r (s i,ai, M). The sum of squares of this error is then averaged (by dividing by N) and multiplied by time T to give the final loss function.
Calculating a predicted Q value and a target Q value: the predicted Q value Q (s i,ai, M; θ) for the current state-action pair s i,ai is calculated using a neural network, and there is already a target Q value Q target(si,ai, M calculated previously.
Calculating an error: then, the error between the predicted Q value and the target Q value is calculated and squared, which ensures that the error is always positive and gives a greater penalty for larger errors.
Calculating an average error: then, the average of the errors of all the data in the empirical playback buffer is calculated, and multiplied by the number of time steps T. This allows for the fact that at different time steps the model may have different sensitivity to errors.
Updating the neural network parameters: finally, a gradient descent or other optimization algorithm is used to minimize this loss function, thereby updating the neural network parameter θ.
The process is a key part of a deep Q learning algorithm, and aims to enable the neural network to better approximate a Q value function, so that a medicine brand popularization system can better select actions, and more effective popularization is achieved.
Example 7: on the basis of the above embodiment, the update formula of the Q value function Q (s, a, M) is:
where α is the learning rate, used to control the updated step size.
Specifically, in Q learning, the purpose of this formula is to update the Q value by taking into account the current and expected future rewards. Here α is the learning rate, which is a super parameter used to control the step size of each update of the Q value. A higher learning rate means that the model will learn more information in a single step update, while a lower learning rate will make the learning process smoother, but may require more time steps to converge. Gamma is a discount factor for determining the importance of future rewards. If γ is close to 1, the model will pay more attention to future rewards, while if γ is smaller, the model will pay more attention to immediate rewards.
Example 8: on the basis of the above embodiment, when the node is a node mapped by medical staff, the node weight a is equal to 0.6; when the node is a node obtained by mapping a patient, the node weight B is 0.4; the spreading factor M is represented using a brand ID of each pharmaceutical brand data.
Specifically, the weight value is a policy for determining the importance of different types of nodes in the propagation network. The node weight of the medical staff is set to 0.6 and the node weight of the patient is set to 0.4, which means that the medical staff is more important than the patient in this popularization strategy. This may be because medical personnel can directly influence the medical decision of more patients, thus giving them a higher weight. The spreading factor M is defined as the brand ID of each medical brand data, meaning that each brand has its own unique identifier, which will help identify and track the different brands during promotion.
Example 9: based on the above embodiment, the calculation formula of the complexity of the propagation network is:
p＝L*(|lg(Pa)|-1)*exp(Do)；
wherein L is the total node number of the propagation network; pa is the number of nodes mapped by the patient; do is the number of nodes mapped by medical personnel.
Specifically, L, which is the total number of nodes of the propagation network, generally represents the scale of the network. The larger the scale, the greater the complexity in general.
The section of the model is obtained through logarithmic transformation and linear transformation based on the number Pa of nodes mapped by the patient. Logarithmic transformation is often used to handle cases where the distribution is extremely non-uniform, enabling the distance between maxima and minima to be reduced, thus making the model more robust.
Exp (Do), which is the mapping of the exponential function to the number of nodes Do mapped by medical personnel. Due to the nature of the exponential function, the complexity p increases rapidly as Do increases. Because medical personnel play a key role in the promotion of pharmaceuticals, their increase can greatly increase the complexity of the network.
In general, this formula combines the total number of nodes of the network, the number of patient nodes and the number of nodes of medical personnel, and calculates the complexity of the propagation network through a series of mathematical transformations. In the background of medicine branding promotion, the complexity can help to know the difficulty of the promotion information in the network, so that the promotion strategy is better designed and adjusted.
Example 10: a method of medical promotion, the method comprising:
Inputting medicine brand data; the medical brand data specifically includes: a pharmaceutical brand name and a brand ID that uniquely identifies the pharmaceutical brand name; marking the brands of the recorded medicines by medical staff, and recording the ages, geographical positions and job positions of the medical staff by each medical staff when marking; the patients score the entered medicine brands, and each patient enters the age of the patient, the geographical position of the patient and the affected symptoms when scoring; firstly, analyzing the scoring result of patients and the scoring result of medical staff of the entered medicine brand data, wherein the method specifically comprises the following steps: screening out medicine brand data of which the patient scoring and the medical staff scoring exceed a set first threshold, screening out medicine brand data of which the patient scoring and the medical staff scoring are lower than a set second threshold, screening out medicine brand data of which the patient scoring exceeds the set first threshold and the medical staff scoring is lower than the set second threshold, or reserving medicine brand data of which the patient scoring exceeds the first threshold as preferable medicine brand data, and reserving medicine brand data of which the patient scoring and the medical staff scoring are in a section formed by the upper limit of the first threshold and the lower limit of the second threshold as less preferable medicine brand data; mapping each patient to a point in a three-dimensional space coordinate system based on the patient age, the patient geographic position and the affected condition corresponding to each patient, mapping each medical staff to a point in the three-dimensional space coordinate system based on the medical staff age, the medical staff geographic position and the medical staff duty of each medical staff, and connecting all points in the three-dimensional space coordinate system as a propagation network; firstly, taking preferable medicine brand data as a transmission factor, randomly selecting an initial node in a transmission network, enabling the transmission factor to be transmitted in the transmission network under the drive of a set transmission model, enabling the number of nodes through which the transmission factor passes to be maximum in a set time interval, recording a transmission path of the transmission factor, taking the transmission path as a brand popularization strategy, carrying out brand popularization, and repeatedly executing the process on less preferable medicine brand data after the preferable medicine brand data are processed, and carrying out brand popularization.
In the description, each embodiment is described in a progressive manner, and each embodiment is mainly described by the differences from other embodiments, so that the same similar parts among the embodiments are mutually referred. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant points refer to the description of the method section.
Those of skill would further appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the various illustrative elements and steps are described above generally in terms of functionality in order to clearly illustrate the interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
The steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. The software modules may be disposed in Random Access Memory (RAM), memory, read Only Memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.
The present invention has been described in detail above. The principles and embodiments of the present invention have been described herein with reference to specific examples, the description of which is intended only to facilitate an understanding of the method of the present invention and its core ideas. It should be noted that it will be apparent to those skilled in the art that various modifications and adaptations of the invention can be made without departing from the principles of the invention and these modifications and adaptations are intended to be within the scope of the invention as defined in the following claims.
Claims (8)
1. A pharmaceutical branding system, said system comprising: the medical brand data input part, the medical staff brand scoring part, the patient brand scoring part, the brand popularization and transmission network construction unit and the brand popularization unit; the medical brand data input part is configured to input medical brand data; the medical brand data specifically includes: a pharmaceutical brand name and a brand ID that uniquely identifies the pharmaceutical brand name; a medical staff brand scoring component configured to provide medical staff with a score for entering a medical brand, each medical staff, when scored, entering a medical staff age, a medical staff geographic location, and a medical staff job; a patient brand scoring component configured to provide patients with a score for the entered medical brands, each patient, when scored, entering a patient age, a patient geographic location, and an affected condition; the brand promotion and propagation network construction unit is configured to analyze the result of the patient scoring of the recorded medicine brand data and the result of the medical staff scoring at first, and specifically comprises the following steps: screening out medicine brand data of which the patient scoring and the medical staff scoring exceed a set first threshold, screening out medicine brand data of which the patient scoring and the medical staff scoring are lower than a set second threshold, screening out medicine brand data of which the patient scoring exceeds the set first threshold and the medical staff scoring is lower than the set second threshold, or reserving medicine brand data of which the patient scoring exceeds the first threshold as preferable medicine brand data, and reserving medicine brand data of which the patient scoring and the medical staff scoring are in a section formed by the upper limit of the first threshold and the lower limit of the second threshold as less preferable medicine brand data; mapping each patient to a point in a three-dimensional space coordinate system based on the patient age, the patient geographic position and the affected condition corresponding to each patient, mapping each medical staff to a point in the three-dimensional space coordinate system based on the medical staff age, the medical staff geographic position and the medical staff duty of each medical staff, and connecting all points in the three-dimensional space coordinate system as a propagation network; the brand promotion unit is configured to take the preferred medical brand data as a transmission factor, randomly select an initial node in a transmission network, enable the transmission factor to be transmitted in the transmission network under the drive of a set transmission model, record the transmission path of the transmission factor when the number of nodes through which the transmission factor passes is maximum in a set time interval, serve as a brand promotion strategy, carry out brand promotion, and repeatedly execute the above-mentioned process on the suboptimal medical brand data after the preferred medical brand data are processed, and carry out brand promotion;
The brand promotion unit takes the optimized medicine brand data as a transmission factor, randomly selects an initial node in a transmission network, and enables the transmission factor to be transmitted in the transmission network under the drive of a set transmission model, so that the method for maximizing the number of nodes through which the transmission factor passes in a set time interval comprises the following steps:
Step S1: defining a state representation: mapping each medical staff member and patient to a point in a three-dimensional spatial coordinate system, the state s being expressed as (x, y, z), wherein x represents age, y represents geographic location, z represents job for medical staff member, and z represents disorder for patient;
Step S2: defining an action space: randomly selecting an initial node in a transmission network, and taking the preferred medicine brand data as a transmission factor M; and propagating under the driving of the set propagation model; defining an action a as a popularization value when a propagation factor M reaches a certain node in a propagation network; defining a set time interval as T; the promotion value is defined as the number of other nodes in the covered circumference area multiplied by the node weight by taking the node as the center and taking the set threshold value as the radius; when the node is a node obtained by mapping of medical staff, the node weight is A; when the node is a node obtained by mapping a patient, the node weight is B;
Step S3: defining a bonus function r (s, a, M): the rewarding function r (s, a, M) measures the effect index of the propagation factor M after the state s executes the action a;
Step S4: defining a Q value function: the Q-value function Q (s, a, M) represents the expected jackpot that the propagation factor M would have obtained after the state s performed action a; estimating a Q-value function using a neural network, denoted Q (s, a, M; θ), where θ is a parameter of the neural network;
Step S5: forming an experience playback buffer in a state-action-rewards sequence collected within a set time T; then, updating a parameter θ of the neural network using a DQN algorithm to minimize a mean square error between the Q-value function and the target Q-value function; at this time, the number of nodes through which the propagation factor passes is the largest.
2. The pharmaceutical branding system of claim 1, wherein said mapping each patient to a point in a three-dimensional spatial coordinate system based on the patient's age, patient geographic location, and condition to which each patient corresponds comprises: taking the age of the patient as an X-axis coordinate, and taking the mean value of longitude and latitude of the geographic position of the patient as a Y-axis coordinate; assigning a uniquely identified condition value to the affected condition as a Z-axis coordinate; the disease values corresponding to different diseases are different.
3. The medical branding system of claim 2 wherein said mapping each medical personnel to a point in a three-dimensional spatial coordinate system based on the medical personnel age, medical personnel geographic location, and medical personnel duty of each medical personnel comprises: taking the age of medical staff as an X-axis coordinate, and taking the mean value of longitude and latitude of the geographical position of the medical staff as a Y-axis coordinate; assigning the medical staff with a uniquely identified duty value as a Z-axis coordinate; the corresponding duty values of different medical staff are different.
4. A pharmaceutical branding system according to claim 3 wherein said target Q function is represented using the formula:
Where s' represents the next state after action a is performed, θ - represents a parameter of the target neural network, and γ is a discount factor for balancing the importance of the instant and future rewards; q target (s, a, M) is a target Q function; a' is the next action; p propagation network complexity.
5. The pharmaceutical branding system of claim 4 wherein in step S5, the parameter θ of the neural network is updated using DQN algorithm to minimize the mean square error between the Q function and the target Q function, the mean square error is expressed using the formula:
Wherein, For mean square error, s i,ai,r(si,ai, M) represents the state-action-reward sequence for a set time T in the empirical playback buffer, N being the size of the buffer; i is a subscript sequence number, the value is a positive integer, and the initial value is 1.
6. The pharmaceutical branding system of claim 5, wherein the updated formula for the Q-value function Q (s, a, M) is:
where α is the learning rate, used to control the updated step size.
7. The pharmaceutical branding system of claim 6, wherein when the node is a node mapped by medical personnel, the node weight a is equal to 0.6; when the node is a node obtained by mapping a patient, the node weight B is 0.4; the spreading factor M is represented using a brand ID of each pharmaceutical brand data.
8. The pharmaceutical branding system of claim 7, wherein the calculation formula for the complexity of the propagation network is:
p＝L*(|lg(Pa)|-1)*exp(Do)；
wherein L is the total node number of the propagation network; pa is the number of nodes mapped by the patient; do is the number of nodes mapped by medical personnel.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310994205.5A CN116883067B (en) | 2023-08-09 | 2023-08-09 | Medical brand popularization system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310994205.5A CN116883067B (en) | 2023-08-09 | 2023-08-09 | Medical brand popularization system and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116883067A CN116883067A (en) | 2023-10-13 |
| CN116883067B true CN116883067B (en) | 2024-10-15 |
Family
ID=88260505
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310994205.5A Active CN116883067B (en) | 2023-08-09 | 2023-08-09 | Medical brand popularization system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116883067B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112053184A (en) * | 2020-08-20 | 2020-12-08 | 腾讯科技（深圳）有限公司 | Promotion information delivery method and device, electronic equipment and storage medium |
| CN113421653A (en) * | 2021-06-23 | 2021-09-21 | 平安科技（深圳）有限公司 | Medical information pushing method and device, storage medium and computer equipment |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA3057521A1 (en) * | 2017-03-23 | 2018-09-27 | Rubikloud Technologies Inc. | Method and system for generation of at least one output analytic for a promotion |
| CN113592520A (en) * | 2020-04-30 | 2021-11-02 | 景德镇陶瓷大学 | Enterprise brand promotion method for marketing |
| US20210365998A1 (en) * | 2020-05-20 | 2021-11-25 | Discovery Communications, Llc | Systems and methods for distributing advertisements for selected content based on brand, content, and audience personality |
| CN113434326B (en) * | 2021-07-12 | 2024-05-31 | 国泰君安证券股份有限公司 | Method and device for realizing network system fault positioning based on distributed cluster topology, processor and computer readable storage medium thereof |
| CN114065914A (en) * | 2021-10-29 | 2022-02-18 | 深圳大学 | Influence maximization seed node set selection method and device |
| CN115130007B (en) * | 2022-08-29 | 2022-11-15 | 深圳市亲邻科技有限公司 | Brand promotion method and system based on user scene positioning |
-
2023
- 2023-08-09 CN CN202310994205.5A patent/CN116883067B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112053184A (en) * | 2020-08-20 | 2020-12-08 | 腾讯科技（深圳）有限公司 | Promotion information delivery method and device, electronic equipment and storage medium |
| CN113421653A (en) * | 2021-06-23 | 2021-09-21 | 平安科技（深圳）有限公司 | Medical information pushing method and device, storage medium and computer equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116883067A (en) | 2023-10-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Walker et al. | Defining uncertainty: a conceptual basis for uncertainty management in model-based decision support | |
| Di Gaspero et al. | The second international timetabling competition (ITC-2007): Curriculum-based course timetabling (track 3) | |
| CN109146741A (en) | A kind of course method for pushing, device and equipment based on capability comparison | |
| Kai et al. | Predicting Student Retention from Behavior in an Online Orientation Course. | |
| CN110415462A (en) | Atm device adds paper money optimization method and device | |
| CN107315775A (en) | A kind of index calculating platform and method | |
| Cohen et al. | A simulation‐based approach in support of project management training for systems engineers | |
| CN108256751A (en) | Online teaching process evaluation method based on big data | |
| Schaumann et al. | Simulating the impact of facility design on operations: A study in an internal medicine ward | |
| Tricahya et al. | Forecasting the amount of pneumonia patients in Jakarta with weighted high order fuzzy time series | |
| CN106558004B (en) | Campus football digital integrated management system | |
| Han | A Fuzzy Logic and Multilevel Analysis‐Based Evaluation Algorithm for Digital Teaching Quality in Colleges and Universities | |
| Gao et al. | Impact of 3D modeling behavior patterns on the creativity of sustainable building design through process mining | |
| Vanli | Ranking of global smart cities using dynamic factor analysis | |
| CN116883067B (en) | Medical brand popularization system and method | |
| Lan et al. | [Retracted] Evaluation System of Music Art Instructional Quality Based on Convolutional Neural Networks and Big Data Analysis | |
| Ntlamelle | The efficacy of SMME incubation as a strategy for enterprise development in South Africa | |
| Deng et al. | An evaluation method of innovation and entrepreneurship education in agricultural and forestry colleges and universities based on elm neural network. | |
| Anupam | Can Digital Games Teach Scientific Inquiry? A Case of How Affordances Can Become Constraints | |
| Zhu et al. | Research on data mining of college students’ physical health for physical education reform | |
| CN117035388A (en) | Risk assessment method and system for target place, terminal equipment and storage medium | |
| Huong et al. | Evaluation of factors related to entrepre-neurial intentions among young pharmacists in the Mekong Delta region: a cross-sectional study in Vietnam. Pharmacia 71: 1–13 | |
| Zheng | [Retracted] Load Prediction Model of Athletes’ Physical Training Competition Based on Nonlinear Algorithm Combined with Ultrasound | |
| Fan et al. | Research on the Application of Decision Tree and Correlation Analysis Algorithm in College Students’ Physical Fitness Analysis | |
| Li | Sports policy and training decision support method based on wireless sensor network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20240909 Address after: Room 206, 2nd Floor, Building 100, Southeast Corner of Xiangjiang Road and Songshan Road Intersection, Yuanhui District, Luohe City, Henan Province 462000 Applicant after: Henan Honeycomb Crowdsourcing Technology Co.,Ltd. Country or region after: China Address before: Room 305, West Building 8, Shuntai Plaza, No. 2000 Shunhua Road, Jinan Area, China (Shandong) Free Trade Pilot Zone, Jinan City, Shandong Province, 250101 Applicant before: Shandong zilue Information Technology Co.,Ltd. Country or region before: China |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |