CN111052124A

CN111052124A - Assessing compliance fidelity of behavioral interventions

Info

Publication number: CN111052124A
Application number: CN201880048021.7A
Authority: CN
Inventors: 托马·本-奇奇; 兰·奇尔卡
Original assignee: Happy Co Ltd
Current assignee: Happy Co Ltd
Priority date: 2017-07-17
Filing date: 2018-07-11
Publication date: 2020-04-21
Also published as: CA3070229A1; EP3655966A1; EP3655966A4; CA3070232A1; WO2019018173A1

Abstract

A computer system apparatus and method performed by such an apparatus for interacting with a user through behavioral interventions aimed at increasing the user's well-being. There are a number of conditions to be met for this behavioral intervention. The method includes receiving input data from the user during the behavioral intervention; performing semantic analysis on at least a portion of the received input data having text to identify terms that satisfy the plurality of conditions; and evaluating a degree of compliance with the behavioral intervention based on a number of integrals satisfying the plurality of conditions. When it is determined that the one or more conditions are not satisfied, the method includes generating an instruction that elicits a reaction from the user that the missing condition is satisfied.

Description

Assessing compliance fidelity of behavioral interventions

Cross Reference to Related Applications

This application claims priority from U.S. provisional patent application No.62/533,423, filed on 17.7.2017, the relevant contents of which are incorporated herein.

Technical Field

The present invention relates to a computing system and a method performed by the system for assessing the magnitude of a user's adherence to a behavioral intervention and for guiding the user to react towards maximum adherence and towards ways of improving the efficacy of the behavioral intervention.

Background

Behavioral intervention typically involves providing a set of instructions to a user/patient and collecting text/language responses. Such interventions have an intended implementation that aims to activate certain psychological mechanisms. The intervention will be effective when the user follows the instructions in the intended mode of implementation. However, when the user is not or only partially in compliance with the intended implementation, the intervention may not be as effective and may not result in more happiness.

For example, a user who writes a negative event would not benefit much from the activity when it is expected to implement a positive event to be written, because the mental mechanism that shifts focus to a positive event would not be activated. As another example, users who are required to mentally depict other people with the same emotion write themselves. Such users do not follow the intended implementation of developing empathetic techniques and do not activate the psychological mechanisms that establish contact with others, resulting in a reduction in the effectiveness of the intervention, and even worse, in a significant reduction in the user's well-being, which is clearly contrary to the intended outcome of the intervention.

In person-to-person conversations (e.g., in one-to-one psychotherapy), a person's compliance with a desired achievement can be assessed and conversed with in a manner that maximizes compliance, thereby increasing effectiveness. In contrast, computer systems typically do not evaluate such methods for compliance with fidelity in software-implemented behavioral interventions. Furthermore, such computer systems do not have a mechanism to direct interaction with a user in a manner that maximizes the desired efficacy by maximizing adherence.

Disclosure of Invention

In view of the foregoing, it is an object of the present invention to provide a computing system/method for assessing the degree to which a user is complying with a behavioral intervention, and reacting in a manner that guides the user towards maximum compliance. It is another object of the present invention to provide a computing system/method for maximizing the enhancement of well-being through behavioral interventions by perfecting adherence to their intended implementation.

According to an embodiment of the present invention, there is provided a computing system for interacting with a user, wherein the computing system initiates a behavioral intervention with the user, the behavioral intervention intended to increase a user's well-being, the behavioral intervention having a plurality of conditions to be satisfied receives input data from the user via at least one sensor during the behavioral intervention, performs a semantic analysis on at least a portion of the received input data having text to determine terms that satisfy the plurality of conditions, and evaluates a degree of compliance with the behavioral intervention based on a quantity of completeness of satisfying the plurality of conditions. The computing system also generates a prompt that is intended to elicit a reaction from the user that is specific to satisfaction of one or more conditions of the plurality of conditions, wherein the plurality of conditions are not satisfied.

As an aspect of this embodiment, the computing system receives input data from a user via at least one sensor during a behavioral intervention to assess a mental state of the user while assessing a degree of adherence to the behavioral intervention.

In another aspect, the computing system evaluates respective degrees of compliance with the behavioral intervention at respective points in time at a plurality of points in time during the behavioral intervention.

As a feature of this aspect, the computing system generates a respective fidelity report at each of the plurality of points in time, the fidelity report containing the respective degree of compliance with the behavioral intervention evaluated at the respective point in time during the behavioral intervention.

As another feature of this aspect, the computing system generates an overall fidelity report for the behavioral intervention at the conclusion of the behavioral intervention based on the plurality of fidelity reports.

As another feature of this aspect, the computing system further includes a display, and at least one of the fidelity report and the overall fidelity report is displayed on the display for viewing by a user. The display of at least one of the fidelity report and the overall fidelity report also enables the user to understand the reasons behind the efficacy of the behavioral intervention.

In another aspect, the behavioral intervention further comprises programmed branch logic for responding to the received input data. The computing system, in the event that it is determined that one or more of the plurality of conditions has not been met, generates a prompt that is intended to elicit a reaction from the user that is specific to the meeting of the one or more of the plurality of conditions, wherein the plurality of conditions are not met, and assigns a priority to the generated prompt such that the generated prompt overrides the programmed branching logic in response to the received input data.

On the other hand, the behavioral intervention is intended to increase the user's well-being.

In another aspect, the behavioral intervention is one of a plurality of activities belonging to a happiness track selected by the user from a plurality of selectable happiness tracks, wherein each happiness track is a distinct program lesson that is intended to enhance the user's happiness.

In another aspect, the behavioral intervention is intended to cause a change in one or more user behaviors.

In another aspect, the received input data includes at least one of spoken and textual data from the user.

In another aspect, the semantic analysis includes pre-training a natural language classifier based on a database of user input data, and the classifier creates one or more labels associated with each of a plurality of conditions.

As a feature of this aspect, the semantic analysis includes determining whether a term identified in the received input data corresponds to one or more tags.

According to another embodiment of the present invention, there is provided a method of interacting with a user through a computing system, wherein the method includes initiating a behavioral intervention with the user, the behavioral intervention aiming to increase a user's happiness, the behavioral intervention having a plurality of conditions to be satisfied: the method includes receiving input data from a user via at least one sensor during a behavioral intervention, performing semantic analysis on at least a portion of the received input data having text to identify terms that satisfy a plurality of conditions, and assessing a degree of compliance with the behavioral intervention based on a number of completeness of the satisfaction of the plurality of conditions, and generating a prompt that is intended to elicit a reaction from the user that is specific to satisfaction of one or more conditions of the plurality of conditions, wherein the plurality of conditions are not satisfied.

As an aspect of this embodiment, the method further includes receiving input data from the user via at least one sensor during the behavioral intervention to assess a mental state of the user while assessing a degree of adherence to the behavioral intervention.

In another aspect, the method further includes evaluating respective degrees of compliance with the behavioral intervention at respective points in time at a plurality of points in time during the behavioral intervention.

As a feature of this aspect, the method further includes generating, at each of the plurality of points in time, a respective fidelity report containing the respective degree of compliance with the behavioral intervention evaluated at the respective point in time during the behavioral intervention.

As another feature of this aspect, the method further comprises generating an overall fidelity report for the behavioral intervention at the end of the behavioral intervention based on the plurality of fidelity reports.

As another feature of this aspect, the method further comprises displaying at least one of the fidelity report and the overall fidelity report on a display for viewing by a user. The display of at least one of the fidelity report and the overall fidelity report also enables the user to understand the reasons behind the efficacy of the behavioral intervention.

In another aspect, the behavioral intervention further comprises programmed branch logic for responding to the received input data. The method includes, upon determining that one or more of the plurality of conditions has not been met, generating a prompt intended to elicit a reaction from a user specific to the meeting of the one or more of the plurality of conditions, wherein the plurality of conditions are not met, and assigning a priority to the generated prompt such that the generated prompt overrides the programmed branching logic in response to the received input data.

On the other hand, the behavioral intervention is one of a plurality of activities belonging to a Happiness track (happy track) selected by the user from a plurality of selectable Happiness tracks, each Happiness track being a distinct program lesson intended to improve the user's Happiness.

According to another embodiment of the present invention, a computing system for interacting with a user is provided, wherein the computing system initiates a empathetic intervention with the user, the empathetic intervention intended to increase a sympathy expression of the user, the empathetic intervention having a plurality of conditions to be satisfied receives input data from the user via at least one sensor during the empathetic intervention, performs a semantic analysis on at least a portion of the received input data having text to identify terms that satisfy the plurality of conditions, and evaluates a degree of compliance with the empathetic intervention based on a number of integrals that satisfy the plurality of conditions. The computing system also generates a prompt that is intended to elicit a reaction from the user that is specific to satisfaction of one or more conditions of a plurality of conditions, wherein the plurality of conditions are not satisfied.

These and other objects, advantages, aspects and features of the present invention are described below and/or will be understood and readily appreciated by those of ordinary skill in the art. Although specific advantages have been enumerated above, various embodiments may include some, none, all, or other technical advantages enumerated, as will become apparent to one of ordinary skill in the art upon reading the following figures and description.

Drawings

FIG. 1 is a block diagram of an exemplary computing system in accordance with the present invention.

Fig. 2 is an exemplary flowchart of an overview of the steps performed by an exemplary embodiment of the present invention.

FIG. 3 is an exemplary diagram of branching logic for an empathetic exercise according to the present invention.

FIG. 4 is an exemplary schematic diagram of a computing system in accordance with the present invention.

Detailed Description

The present invention is directed to a computing system and a method employed by a technical apparatus that provides an environment for interacting with a (human) user through behavioral interventions, and during such interactions, the degree of adherence to the user's behavioral interventions is evaluated. The computing system uses a variety of sensors and analytical techniques to perform compliance fidelity assessments at key steps of the intervention and ultimately develops a personalized and/or overall fidelity report that is the basis for understanding how certain behavioral interventions are or are not functional and why certain behavioral interventions are or are not functional.

It should be understood at the outset that although exemplary embodiments are illustrated in the figures and described below, the principles of the present disclosure may be implemented using any number of techniques, whether currently known or unknown. The present disclosure should in no way be limited to the exemplary implementations and techniques illustrated in the drawings and described below.

As used herein, the term "behavioral intervention" or simply "intervention" is intended to be broadly construed, and thus, the term may include various interventions specifically designed to increase the physical and mental health of a user/patient. According to the present invention, an "intervention" may simply be an activity based on a previous evidence-based study, indicating that when someone engages in the activity (as expected), that person benefits in terms of their psychological and/or physical well-being. According to the present invention, the computing system "provides" intervention to the user. In general, the term is intended to mean that the computing system loads an intervention, i.e., a stored executable or mobile application, and initiates and/or engages the user in a set of activities. Interventions generally consist of a set of pre-arranged activities or sessions or tasks that are deployed or performed between a user or user and a trainer (or virtual trainer). Interventions also generally have the purpose of activating the user's mind and/or certain psychological or physiological mechanisms within the body by driving certain emotional reactions of the user. Thus, interventions usually carry with them the intended implementation, i.e. the development of a method for the set of pre-arranged activities, which is performed by the creator of such interventions to most effectively achieve the potential objectives behind the intervention. The contemplated implementations may come in the form of criteria, conditions, requirements, or factors, each of which is designed to be satisfied by a user by performing a particular action or saying a particular word. Thus, the most desirable and effective way to perform intervention is to have the user faithful to the intended implementation during the intervention.

According to various embodiments of the invention described herein, intervention may be used to train a user to develop certain skills or to change certain habitual behaviors to address issues faced in the user's life. For example, such interventions may include behavioral modification interventions, active interventions, and clinical interventions (e.g., Cognitive Behavioral Therapy (CBT), Acceptance and Commitment Therapy (ACT), focus resolution therapy SFT), Behavioral Activation (BA), or behavioral modification interventions). Further in accordance with the present invention, such interventions are of variable length, in that the computing system dynamically decides how to continue interaction at each alternation of interventions based on an assessment of the user's compliance with the intended implementation, as will also be described herein.

Referring now to the drawings in which like numerals represent the same or similar elements and initially to FIG. 1, a computing system 100 configured in accordance with the present invention is illustratively shown in accordance with one embodiment. Computing system 100 includes one or more processors 110, and one or more processors 110 process various input data and stored data and control the operation of other components within computing system 100 to enable "behavioral intervention" between one or more users 200 and computing system 100 as described herein. As will be further described, the processor 110 processes the data by performing a number of mathematical algorithms and analytical calculations. The processor 110 may also be a plurality of processing units, each performing a respective mathematical algorithm and/or analytical calculation. In some embodiments, the processor 110 is enhanced by artificial intelligence.

The computing system 100 also includes a plurality of sensors 120. The plurality of sensors 120 may include speakers/microphones, still image cameras, moving image cameras, biometric sensors, and the like. Each sensor 120 is configured to obtain user input data, and each sensor may further include one or more respective processing units to process the obtained input data in conjunction with the processor 110. Computing system 100 also includes an interface 130 that allows user 200 to operate the computing system and display 140 to present information to user 200. In some embodiments, the interface 130 and the display 140 may appear as one unit, such as a touch screen display.

Computing system 100 also includes communication units/devices 150, input/output ports 160, and memory 170. The communication unit/device 150 allows the computing system 100 to communicate with other electronic devices of the user within the user 200 accessory or with additional sensors over the network 300. Network 300 may include wireless communications, wired communications, and the like. The network 300 may include the internet, a wide area network, a local area network, or the like. The computing system 100 may use the I/O ports 160 for inputting and outputting data. The computing system 100 also includes memory 170 that stores programs and applications. The memory 170 may store intervention data or may store interventions retrieved from a server 400 having an intervention database in the server 400 locally.

Computing device 100, as well as other electronic devices of users or additional sensors, may be part of network 300 or connected to network 300 and coupled to server or service provider 400.

The dashed lines in FIG. 1 indicate that user 200, network 300, server 400, and computing system 100 may be directly, indirectly, or remotely connected to any one or more of user 200, network 300, server 400, or computing system 100 via a communication path. One or more of computing system 100, network 300, and server 400 may reside on a single computer, be distributed over multiple computers, or be partially or fully internet based.

According to certain exemplary embodiments of the present invention, the computing system is embodied as an aggressive psychology service referred to herein as "happy". "happy" is a novel, science-based online service for participation, learning, and training happiness skills. Happify is based on a framework developed by psychologists and researchers in a range of therapeutic disciplines (e.g., CBT, mindsets, aggressive psychology, etc.), and may help users develop certain skills related to Happify, such as taste, thank you, craving, giving and sympathy (or s.t.a.g.e.)^TM). In some embodiments, each skill is developed using various activities, ordered incrementally by skill level, that progressively unlock as the user progresses to establish the skill. Using happefy, a user can select a "track" containing a series of activities that are intended to address a particular life condition or goal.

The happy system may be implemented on a user's mobile electronic device (e.g., a smartphone or tablet computer) or may be implemented on a user's Personal Computer (PC). Happefy may be embodied within a mobile application program, executable software program, or other suitable form. For example, a user may download and install a mobile application that provides a happy service. The user selects a "happiness" track through the mobile application and provides it with a series of activities directed at improving the user's happiness in accordance with the selected track.

Several patents listed below describe the Happify system and the operation of the Happify system in detail, with specific patents: U.S. patent application No.14/284,229 entitled "SYSTEMS AND METHODS FOR PROVIDING ON-LINE SERVICES", U.S. patent application No.14/990,380 entitled "DYNAMIC INTERACTION SYSTEM AND METHOD", and U.S. patent application No.15/974,978 entitled "SYSTEMS AND METHODS FOR DYNAMIC USER INTERACTION FOR IMPROVINGHAPPNESS", the entire contents of each of which are hereby incorporated by reference. For brevity, no further details regarding the happefy system are provided herein (unless otherwise stated herein).

In accordance with the present invention, an exemplary computing system embodying the Happify system provides a user with a set of activities as part of a selected happiness track. These "activities" may also be referred to herein as another form of "intervention". Each activity has its own intended implementation, purpose, and scientific-based basis in developing skills related to increasing a user's well-being, and the computing system employs various computer-specific mechanisms to track and assess a user's adherence at each activity as each activity is provided to and operated by the user.

An overview of the steps performed by an exemplary computing system according to the present invention is shown in FIG. 2.

According to the invention, step S201 requires interaction with the user in an iterative manner (i.e. dialog by text or by speech). For example, iterative interactions initiated by the computing system may include providing prompts to the user, receiving input data from the user, providing subsequent prompts to the user, receiving other input data from the user, and so forth.

Step S202 entails collecting data from an array of sensors that extract features from the user' S response at key steps of the interaction. For example, the computing system may communicate (e.g., wired or wireless) with one or more devices configured to collect user information, such as a camera, speaker, microphone, thermal sensor, motion sensor, fingerprint detector, keyboard, and so forth. Such devices may include various structures and/or functions, and may also include one or more processors to execute various natural language understanding tools.

Step S203 entails using the collected data to assess a user' S level of compliance with one or more anticipated implementations of the provided activity at a critical step of the interaction. For example, any given behavioral intervention may include a number of components, conditions, and/or criteria needed to treat the intervention as complete. The computing system performs an analysis that will allow the computing system to identify whether or how certain components, conditions, and/or criteria are being processed or are being processed at a critical step of the interaction.

Step S204 entails using the assessment in each key step of the interaction/dialog to determine the best reaction to the user to guide the interaction towards maximized or improved adherence. For example, while a given interaction may include a predetermined sequence or course of action, based on the evaluation, the computing system may intervene or deviate from the predetermined sequence, for example, for the purpose of guiding the user towards maximized or improved adherence, by asking different questions, for example, using programmed branching logic.

Step S205 entails evaluating the overall degree of compliance with the anticipated implementation of the intervention at the end of the interaction.

Finally, step S206 entails presenting the overall level of compliance for each intervention to the user as a score through the behavioral intervention program and using the overall level of compliance as an indication of the program.

Additional details of each of the above steps will be discussed in greater detail herein.

Initially, as described above, a computing system in accordance with the present invention provides behavioral intervention to a user through iterative interaction with the user. As described herein, such interaction includes initiating a dialog with the user, assessing the user's current mental state, providing an activity or task that the user will perform, and so forth. In accordance with the present invention, a computing system receives input data either directly from a user or indirectly via one or more sensors, analyzes the input data, and responds back to the user. This iterative interaction continues until, for example, a desired result is obtained.

In one or more embodiments, a computing system may be equipped with specific software that enables the computing system to interact with a user dynamically in response to ongoing input data, or emotionally (e.g., emphatically) for a variety of reasons.

According to embodiments of the present invention, a computing system is provided with novel capabilities to simultaneously, sequentially or independently assess the fidelity of compliance of a user at various steps during an intervention. For example, in one embodiment, a computing system may collect and analyze user input text/language data to continuously update a user's mental state while performing analysis on the collected input text/language data to assess the user's compliance fidelity each time input data is received. In another embodiment, the computing system evaluates the user's compliance fidelity before or after performing other analysis on the input text/language data for each input datum. In another embodiment, the assessment of fidelity may be performed independently at each step of the intervention or at predetermined time intervals throughout the intervention.

As described herein, a computing system employs sensors configured to collect and analyze user input data. The term "sensor" as used herein includes a computer keyboard or other type of computer data input device, such as a touch screen, mouse, microphone, etc., as well as other devices disclosed herein or known in the art by which a user may actively or passively provide information to the system. Various types of sensors may be employed to collect audible or visual data, or a user may directly type or write input data received by the computing system. The sensors not only collect data, but also perform analyses, below a list of exemplary analysis techniques performed by the sensors with the purpose of extracting information from the input data and based on such information, assessing the user's degree of compliance with the intended implementation of the behavioral intervention.

I. Named entity recognition

One or more sensors as described herein are equipped with a processing unit to perform a "named entity recognition" analysis, which is the ability to recognize entities in the body of text and refer to the recognized entities in a unified, normalized form, regardless of the specific language. For example, the phrases "Facebook originator", "Mark zakherg" and "Facebook originator" all refer to the same entity whose presence in the text can be detected. According to the invention, the computing system uses this analysis to detect certain entities that the user mentions during the intervention.

For example, in a new exercise, the user may be asked to describe his or her romantic life. The user may already have multiple named entities stored in the computing system, such as one named entity at work for his or her owner to use, and another named entity stored for the spouse. These named entities may have been detected and stored in previous sessions. In this exercise, the user provides a reaction, but the computing system detects the entity "boss". Thus, the computing system does not proceed to the next prompt in the intervention, but rather maximizes adherence by the drum user re-transferring the point of interest into his or her romantic life, thereby increasing adherence fidelity.

Exemplary adherence cues that the computing system presents in response to the named entity "boss" detected in the "romantic life" exercise are:

TABLE 1

Then, if "boss" entities are detected continuously, the resulting adherence score for the resulting intervention will be low.

Other details of the details of this technique have been omitted herein for the sake of brevity. The following list is an exemplary publication specifically describing this technology, which is incorporated herein by reference: nadeau, David&Sekine, Satoshi, "A surveyed inventory recognition and classification," Lingvicine investigations 30.1(2007): 3-26; tjong Kim Sang, e.f.&"Introduction to the coding L-2003shared task" of De Meulder, F., Language-independent resource registration ", Proceedings of the seven-level conference on Natural Language learning at HLT-NAACL, Vol.IV., pp.142-; and wikipedia:https://en.wikipedia.org/wiki/ Named-entity_recognition。

II.text pattern matching

The processing units of the one or more sensors described herein may also perform "text pattern matching" analysis, which refers to the ability to detect the presence of certain string patterns (character sequences) in the body of text by matching these patterns to the given text. The pattern is typically provided using regular expressions. For example, the pattern "×" matches all strings of all lengths that contain the letter "b".

In accordance with the present invention, a computing system may use pattern matching to detect certain words or phrases that indicate an increase or decrease in fidelity of adherence to an intended implementation of an intervention. For example, pattern matching may be used to detect derogative and negative words in thank you intervention, where the implementation is expected to have a positive and relaxed mood. When derogative and negative words are detected, the user will be guided to try to use a more aggressive mood and avoid using that language, resulting in an improved adherence to fidelity score.

Other details of the details of this technique have been omitted herein for the sake of brevity. The following list is an exemplary publication specifically describing this technology, which is incorporated herein by reference: knuth, Donald E., James H.Morris, Jr, and Vaughan R.Pratt, "Fast pattern matching in strings", SIAM journal on computing6.2(1977): 323-350; knuth, d.e., Morris, jr., j.h.&Pratt's "Fast pattern matching instrings”V.R.,SIAM journal on computing,6(2),323-350(1977)；Navarro,G.&"Flexible pattern matching in sequences" by Raffinot, M.A. practical on-line search algorithms for texts and biological sequences ", Cambridge university Press (Cambridge university Press), 2002; and wikipedia:https://en.wikipedia.org/wiki/Pattern_ matching。

III.emotion analysis and emotional tone

Sentiment analysis is another technique that may be performed by a sensor. Basic sentiment analysis can identify sentiment polarity between "negative" and "positive" in text. More advanced emotion analysis can identify specific emotions such as "sadness", "anger and" happiness "in the text. Furthermore, sentiment analysis of the text may identify which segments (i.e., particular words) in the text are indicative of the detected emotion. When a conversation is conducted through speech rather than written text, the emotional tone of the text may be further recognized by recognizing acoustic features associated with different emotions.

Some intervention according to the invention requires users to describe negative ideas that plague them. The user responds with positive thoughts, and in emotion analysis, when such responses are detected, low compliance fidelity is manifested. The computing system then responds in an attempt to guide the user to think about the negative ideas they want to solve and overcome, rather than the positive ones. If the user eventually fails to comply, the compliance score for the intervention remains low after the intervention is complete.

Other details of the details of this technique have been omitted herein for the sake of brevity. The following list is an exemplary publication specifically describing this technology, which is incorporated herein by reference: "Opinion accounting and sensory analysis", Pang, Bo, and Lillian Lee, reasons and regulations

in InformationRetrieval 2.1–2(2008):1-135；“Survey on speech emotion recognition:Features,classification schemes,and databases”，El Ayadi,Moataz,Mohamed S.Kamel,andFakhri Karray,Pattern Recognition 44.3(2011): 572-587; "Analysis of organizational using interface expressions, speed and multimodal information", Busso, Carlos, et al, Proceedings of the 6th international conference on multimodal interfaces, ACM, 2004; and (4) google: https:// group.google.com/natural-language/; IBM Watson:

https://cloud.google.com/natural-language/&_

https://www.ibm.com/watson/developercloud/tone-analyzer.html；

microsoft:

https://www.microsoft.com/reallifecode/2015/11/29/emotion-detection- and-recognition-from-text-u sing-deep-learning(ii) a And wikipedia:https:// en.wikipedia.org/wiki/Sentiment_analysis。

IV semantic analysis

"semantic analysis" refers to the various capabilities of associating words, phrases, sentences, and paragraphs with the entire text as a whole. Sensors according to the present invention may use semantic analysis to assess compliance fidelity. For example, some intervention may require the user to describe the challenges he/she is facing in the work. By using a latent dirichlet allocation topic model pre-trained on other textual data, the computing system may recognize that the user is discussing two main topics: "vacation and leisure travel" and "summer". Thus, the computing system concludes that the user is describing a summer vacation experience rather than a work challenge and requires the user to focus attention on the work challenge so that the user is more in line with the intended implementation of the intervention.

Other examples of semantic analysis include:

"Part-of-speed tagging", Voutilainen, atom, The Oxford handbook of computational rules (2003): 219-; "Laten Semantic Analysis", Landauer, Thomas K, John Wiley & Sons, Ltd, 2006; and "tension dirichlet allocation", Blei, DavidM., Andrew Y.Ng, and Michael I.Jordan, Journal of machine Learning research3.Jan (2003): 993-. Each of the above publications is incorporated by reference herein in its entirety.

Classification of natural language

The "natural language classification" technique assigns text to one of a finite number of classes. Categories are typically defined by tags and can decide which tags to use. An exemplary use of this technique according to the present invention is as follows to detect a degree of compliance that a user exhibits with an intended implementation.

TABLE 2

In this example, the user is required to attend to another person and present a concourse to them. Natural language classifiers have been pre-trained and can be classified between two classes: 1) self-concentration (author is concentrated on itself); 2) focus on others (the author focuses on others). The user answers "i think that joe does not like me" and upon sending this text to the classifier, the classifier returns the label "self-care" indicating that the user is not concentrating on others, resulting in lower compliance fidelity. The system may then respond by encouraging the user to describe the thing from the perspective of others to improve compliance fidelity.

Other details of the details of this technique have been omitted herein for the sake of brevity. The following list is an exemplary publication specifically describing this technology, which is incorporated herein by reference: IBM Watson:

https:// www.ibm.com/watson/developergroup/nl-classifier.html; "Malettext classification software" http:// mallet.cs. um. ass. edu/classification. php; "extreme of text classification algorithms", Aggarwal, Charu C., and ChengXiangZhai, Mining text data (2012): 163-222; and wikipedia:https://en.wikipedia.org/ wiki/Document_classification。

the list of the aforementioned analysis techniques is not exhaustive, but merely exemplary. Exemplary embodiments may also use other unsupervised analysis techniques, such as topic modeling, to extract potential tags for text classification. Any of the techniques disclosed herein may be performed by a processing unit within the respective sensor, or may be performed within one or more processing units external to the respective sensor.

An example will now be described with reference to fig. 3, in which a computing system in accordance with the present invention initiates empathetic intervention with a user. The intended implementation of this intervention is to let the user "walk with someone else's shoes". An exemplary dialog is as follows:

TABLE 3

To accomplish the intended implementation, this particular intervention requires adherence to the following four requirements or conditions:

1) focusing on another person rather than on himself.

2) One situation in the life of the opponent is described, as well as the field of life of such a situation (for example, the situation is "coping with diabetes", where the field of life is "healthy").

3) At least one positive feature of another person is mentioned.

4) Describing the emotion another person is experiencing.

All these requirements are related to the aim of proving that the efficacy of the sympathy and intervention is maximized when all the requirements are fulfilled. Thus, upon completion of each of these four requirements, the user will obtain a partial score of, for example, 25 points. If all four requirements are met, the user will get a full score of 100. Each step of the intervention may be scored and these scores may be presented to the user at the end of the intervention in the form of a personalized fidelity report or an overall fidelity report. An important purpose of presenting fidelity reports to the user is to let the user know why certain interventions work or do not work, why certain interventions work better than others and/or how the user improves the efficacy of a given intervention. Referring to FIG. 3, the computing system, for example, assigns "Y" when one or more of these conditions are met, and assigns "N" or "-" when one or more of these conditions are absent. In this example, the ideal response is assigned [ YYYY ].

Returning to this example, the following dialog further proceeds:

TABLE 4

The user's first reaction ("joe") is analyzed against programmed branching logic and assigned [ YNNN ] because the computing system only detects the first of four conditions. The computing system determines that there are more conditions to satisfy and the dialog continues. Since these four conditions were not detected, the second response of the user ("i am a good person") was analyzed and assigned [ NNNN ]. The computing system determines that there are more conditions to satisfy and that the user is deviating from the topic. The computing system responds in a manner that brings the user back to track. Finally, since all four conditions ("joe", "tough", "heavy work", "feeling awkward") were detected, the third reaction of the user (joe is a very tough person, is dealing with heavy work, and feels awkward) was analyzed and assigned [ yyyyy ]. The computing system determines that all conditions of intervention have been met and ends the dialog.

According to the invention, it is not just a matter of sequentially asking a stack of different questions until the user is fully compliant. Rather, in essence, the present invention is specific to each subsequent prompt to guide the user to achieve maximum adherence.

In other words, at each step of the intervention, the computing system evaluates the user-provided (e.g., textual) input data and, based on the evaluation, adjusts the next prompt accordingly to guide the user to maximized adherence. Thus, the next hint may override or override the pre-arranged next in-line hint based on behavioral intervention.

In some embodiments, the intervening programming logic is initially designed such that the computing system executes inputs on each branch logic, for example, when "a" → going to "X", when "B" → going to "Y", when "C" → going to "Z", and so on. However, rather than simply detecting the presence of "A" or "B" or "C", the computing system may also investigate the extent or amount to which "A" or "B" or "C" is performed in accordance with the present invention. Thus, the computing system may not branch directly to "X" or "Y" or "Z," but rather guide the user in a direction that will first maximize adherence to "a" or "B" or "C" of behavioral intervention.

Returning to the sympathy intervention example, in the portion of the intervention that requires the user to list the positive traits of another person, the computing system parses the statement entered by the user to determine whether the statement contains terms that belong to the "positive traits" category. For example, a computing system performs semantic analysis (e.g., LDA topic modeling) on user text throughout a database to perform co-estrus exercises and identify terms that people use to describe other people's positive features. For example, the terms "smart", "strong" and "friendly" are generally identified as describing positive traits. The identified terms are then used as labels to train a natural language classifier that will identify whether a given text can be classified as one of the three identified classes. Phrases such as "he is clever," "she is clever," or "this is a very clever person," etc., may be added to tags identified as "clever" classes according to one or more of the above exemplary analysis techniques.

For additional details regarding modeling of LDA topics, see "late dichotomy allocation" by Blei, D.M., Ng, A.Y., and Jordan, M.I., Journal of Machine Learning Research, 3 (month 1), 993-. Each of these references is incorporated herein by reference in its entirety.

If the statement entered by the user fails to contain one or more terms that the computing system believes correspond to a positive trait of another, the computing system notifies the user of the failure and requires the user to try again. If the user continuously misses one or more particular conditions for a given intervention, the computing system may adjust subsequent prompts to remedy the deficiency. If the user deviates from the intended implementation, the computing system will adjust the prompt appropriately to return the user to normal. In some embodiments, the prompts may be specifically fine-tuned to maximize compliance, while in some other embodiments, the computing system may more directly guide the user to achieve maximum compliance.

Another exemplary conversation for co-emotional behavioral intervention proceeds as follows:

TABLE 5

In this example, the programmed branching logic returns [ YNN ] and [ YYNY ] for the user's first and second reactions, respectively. The computing system detects that satisfaction of "other concerns", "positive traits", and "emotions" has been met, but the user has not complied with the "realm" requirements of this intervention. Thus, the interventions are dynamically changed in order to place particular emphasis on the user satisfying the conditions of loss and achieving maximum adherence.

Another example will be described with reference to fig. 4. In this example, during a dialog between the computing system and the user, the following steps are described:

1) the user types or speaks.

2) The user's text and/or speech signals are "decoded" (i.e., the input data is run through a plurality of sensors) by a decoder component of the computing system.

3) The results created and stored by the computing system are a rich description of the user that provides information such as sentiment mood, sentiment, semantics, etc. (e.g., based on the output of one or more sensors as described herein).

4) The dialog manager then analyzes the sensor output, as well as the context of the dialog (i.e., representing the interaction with the user in the current dialog) and the broader context in the dialog, both of which are stored in a repository of context variables.

5) Based on this analysis of the decoded user input and context, the dialog manager then determines how to respond next and combines the appropriate responses to the user with the intent of causing some user input in the next round.

6) The dialog manager also updates the repository of context variables (within and between dialog context variables).

According to an embodiment of the invention, the dialog manager follows the above-described steps in order to maximize the user's compliance with the intended implementation of the intervention.

The appearances of the phrase "in one embodiment" or "in an exemplary embodiment" or any other variations of the phrase in various places throughout the specification are not necessarily all referring to the same embodiment, but rather are merely meant to imply that a particular feature, characteristic, structure, or the like described in connection with the embodiment described is included in at least one embodiment.

The techniques described herein may be incorporated into systems, methods, and/or computer program products comprising a non-transitory computer-readable storage medium having computer-readable program instructions to cause a processor to perform aspects of one or more embodiments.

The program instructions are computer-readable and may be downloaded from a computer-readable storage medium to one or more computing/processing devices via a network, or may be downloaded to an external computer or external storage device via a network, which may include a local area network, a wide area network, a wireless network, or the internet.

Additionally, the network may include wireless transports, routers, firewalls, switches, copper transmission cables, optical transmission fibers, edge servers, and/or gateway computers. In the respective computing/processing device, the network adapter card or network interface in each computing/processing device receives computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in the computer-readable storage medium.

As used herein, a computer-readable storage medium should not be interpreted as a transitory signal, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium, or an electrical signal transmitted through an electromagnetic wave. The computer readable storage medium may be, for example, but is not limited to, magnetic, electronic, optical, semiconductor, electromagnetic, or any suitable combination of the foregoing, and may be a tangible device that can retain and store instructions for use by an instruction execution device.

The following is a list of more specific examples, but not exhaustive, of computer-readable storage media: punch cards, raised structures in grooves, or other mechanical coding devices having instructions recorded thereon, erasable programmable read-only memory, static random access memory, portable compact disc read-only memory, digital versatile disc, portable computer diskette, hard disk, random access memory, read-only memory, flash memory, memory stick, floppy disk, and any suitable combination of the foregoing.

The operations of one or more embodiments described herein may be performed by program instructions that may be machine instructions, machine-related instructions, microcode, assembly instructions, instruction set architecture instructions, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including an object oriented programming language such as, but not limited to, C + +, and other conventional procedural programming languages.

As will be apparent to those skilled in the art from this description, the program instructions may have the capability to execute entirely on the user's computer, partly on a remote computer, partly on the user's computer, entirely on the remote computer or server, or as a stand-alone software package. In an arrangement in which the "executes entirely on the remote computer or server," the remote computer may be connected to the user's computer through any type of network, including a wide area network or a local area network, or the connection may be made to an external computer.

In some embodiments, an electronic circuit comprising, for example, a field programmable gate array, a programmable logic circuit, or a programmable logic array, may execute program instructions by personalizing the electronic circuit with state information for the program instructions in order to perform aspects of one or more embodiments described herein.

These program instructions may be stored in a computer-readable storage medium that can direct a computer, a programmable data processing apparatus, and/or other devices to function in a particular manner, such that the computer-readable storage medium having stored therein the instructions which comprise instructions which implement the aspects of the function/act specified in the flowchart and/or block diagram block or blocks. These program instructions may also be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions/acts specified in the flowchart and/or block diagram block or blocks.

The computer-readable program instructions may also be loaded onto a computer, other programming means, or other devices to produce a computer-implemented process such that the instructions which execute on the computer, other programming means, or other devices implement the functions/acts specified in the flowchart and/or block diagram block or blocks.

The block diagrams in the figures and/or other figures and/or flowcharts illustrate the functionality, architecture, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present invention. In this regard, each block in the block and/or other diagrams and/or flowchart illustrations may represent a module, segment, or portion of instructions, which comprises one or more executable instructions for implementing the specified logical function(s). In some alternative embodiments, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block and/or other diagrams and/or flowchart illustration, and combinations of blocks in the block and/or other diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

Modifications, additions, or omissions may be made to the systems, devices, and methods described herein without departing from the scope of the disclosure. For example, components of the system and the device may be integrated or separated. Moreover, the operations of the systems and devices disclosed herein may be performed by more, fewer, or other components, and the methods described may include more, fewer, or other steps. Additionally, the steps may be performed in any suitable order. As used herein, "each" refers to each member of a set or each member of a subset of a set. To assist the patent office and any reader of any patent issued in accordance with this application in interpreting its appended claims, applicants desire to note that applicants do not intend to use any appended claims or claim elements to refer to 35u.s.c. § 112(f), unless "method for … …" or "step for … …" is explicitly used in a particular claim.

In view of the foregoing disclosure, the computing systems and techniques of the present invention have been described for interacting with a user. In accordance with the disclosure provided herein, a computing system uses behavioral interventions to interact with a user to improve well-being or more generally to alleviate or alleviate symptoms of mental health conditions, such as depression and anxiety, such interactions requiring assessment of compliance fidelity to the behavioral interventions by the computing system to maximize the efficiency of the behavioral interventions. In further accordance with the disclosure provided herein, a computing system evaluates compliance fidelity and dynamically adjusts the prompts during behavioral interventions to guide the user in achieving maximized compliance.

Claims

1. A computing system for interacting with a user, the computing system comprising:

at least one processor;

at least one sensor;

at least one memory storing executable software that, when executed by the at least one processor, causes the at least one processor to:

initiating a behavioral intervention on a user, the behavioral intervention intended to increase the user's well-being, the behavioral intervention having a plurality of conditions to be satisfied;

receiving input data from the user via the at least one sensor during the behavioral intervention;

performing semantic analysis on at least a portion of the received input data having text to determine terms that satisfy the plurality of conditions; and

evaluating a degree of compliance with the behavioral intervention based on a quantity of completeness at which the plurality of conditions are satisfied,

wherein the executable software stored in the at least one memory is adapted to cause the at least one processor to generate a prompt intended to elicit a reaction from a user specific to satisfaction of one or more conditions of a plurality of conditions, wherein the plurality of conditions are satisfied.

2. The computing system of claim 1, wherein the executable software stored in the at least one memory is further adapted to cause the at least one processor to receive input data from a user via the at least one sensor during a behavioral intervention to assess a mental state of the user while assessing a degree of adherence to the behavioral intervention.

3. The computing system of claim 1, wherein the executable software stored in the at least one memory is further adapted to cause the at least one processor to evaluate, at a plurality of points in time during the behavioral intervention, respective degrees of compliance at respective points in time with the behavioral intervention.

4. The computing system of claim 3, wherein the executable software stored in the at least one memory is further adapted to cause the at least one processor to generate a respective fidelity report at each of the plurality of points in time, the fidelity report containing the degree of compliance with the behavioral intervention assessed at the respective point in time during the behavioral intervention.

5. The computing system of claim 4, wherein the executable software stored in the at least one memory is further adapted to cause the at least one processor to generate an overall fidelity report for the behavioral intervention based on a plurality of fidelity reports at the end of the behavioral intervention.

6. The computing system of claim 5, further comprising:

a display device, which is used for displaying the image,

wherein the executable software stored in the at least one memory is further adapted to cause the at least one processor to display at least one of the fidelity report and the overall fidelity report on the display for viewing by the user, and wherein the display of at least one of the fidelity report and the overall fidelity report further enables the user to understand reasons behind the efficacy of the behavioral intervention.

7. The computing system of claim 1, wherein the behavioral intervention further comprises programmed branching logic for responding to the received input data, an

Wherein the executable software stored in the at least one memory is further adapted to cause the at least one processor, upon (i) determining that one or more of the plurality of conditions has not been met; and in

(ii) Generating a prompt that is intended to elicit a reaction from a user that is specific to satisfaction of one or more conditions of the plurality of conditions, wherein the plurality of conditions are not satisfied,

generating hints for assigning priority such that the generated hints override the programmed branching logic in response to received input data.

8. The computing system of claim 1, wherein the behavioral intervention is intended to increase a user's happiness.

9. The computing system of claim 1, wherein the behavioral intervention is one of a plurality of activities belonging to a Happiness track (happy track) selected by the user from a plurality of selectable Happiness tracks, wherein each Happiness track is a distinct program lesson that is intended to improve the user's Happiness.

10. The computing system of claim 1, wherein the behavioral intervention is intended to cause a change in one or more user behaviors.

11. The computing system of claim 1, wherein the received input data comprises at least one of verbal and textual data from the user.

12. The computing system of claim 1, wherein the semantic analysis comprises pre-training a natural language classifier based on a database of user input data, and wherein the classifier creates one or more labels associated with each of a plurality of conditions.

13. The computing system of claim 12, wherein the semantic analysis further comprises determining whether a term identified in the received input data corresponds to the one or more tags.

14. A method for interacting with a user via a computing system, the method comprising:

evaluating a degree of compliance with the behavioral intervention based on a number of integrals satisfying the plurality of conditions; and

a prompt is generated that is intended to elicit a reaction from a user that is specific to satisfaction of one or more conditions of a plurality of conditions, wherein the plurality of conditions are not satisfied.

15. The method of claim 14, further comprising:

receiving input data from a user via the at least one sensor during a behavioral intervention to assess a mental state of the user while assessing a degree of adherence to the behavioral intervention.

16. The method of claim 14, further comprising:

evaluating, at a plurality of points in time during the behavioral intervention, respective degrees of compliance with the behavioral intervention at respective points in time.

17. The method of claim 16, further comprising:

generating, at each of the plurality of time points, a respective fidelity report containing the respective degree of compliance with the behavioral intervention evaluated at the respective time point during the behavioral intervention.

18. The method of claim 17, further comprising:

generating an overall fidelity report for the behavioral intervention based on a plurality of fidelity reports at the end of the behavioral intervention.

19. The method of claim 18, further comprising:

displaying at least one of the fidelity report and the overall fidelity report on a display for viewing by the user,

wherein the display of at least one of the fidelity report and the overall fidelity report further enables the user to understand reasons behind the efficacy of the behavioral intervention.

20. The method of claim 14, wherein the behavioral intervention further comprises programmed branch logic for responding to the received input data, an

The method further comprises the following steps:

upon (i) determining that one or more of the plurality of conditions has not been met; and

upon (ii) generating a prompt that is intended to elicit a reaction from the user that is specific to satisfaction of one or more conditions of the plurality of conditions, wherein the plurality of conditions are not satisfied,

21. The method of claim 14, wherein the behavioral intervention is intended to increase user well-being.

22. The method of claim 14, wherein the behavioral intervention is one of a plurality of activities belonging to a Happiness track (happy track) selected by the user from a plurality of selectable Happiness tracks, wherein each Happiness track is a distinct program lesson aimed at improving the user's Happiness.

23. The method of claim 14, wherein the behavioral intervention is intended to cause a change in one or more user behaviors.

24. The method of claim 14, wherein the received input data comprises at least one of verbal and textual data from the user.

25. The method of claim 14, wherein the semantic analysis comprises pre-training a natural language classifier based on a database of user input data, and wherein the classifier creates one or more labels associated with each of a plurality of conditions.

26. The method of claim 25, wherein the semantic analysis includes determining whether a term identified in the received input data corresponds to the one or more tags.

27. A computing system for interacting with a user, the computing system comprising:

at least one processor;

at least one sensor;

initiating an empathetic intervention on a user, the empathetic intervention aiming to increase empathetic expression of the user, the empathetic intervention having a plurality of conditions to be met;

wherein the executable software stored in the at least one memory is adapted to cause the at least one processor to generate a prompt that is intended to elicit a reaction from a user that is specific to satisfaction of one or more conditions of a plurality of conditions, wherein the plurality of conditions are not satisfied.