EP2160882A1 - Method and device for establishing relation between pairs in a network - Google Patents

Abstract

The method of the invention is used for establishing a relation between pairs in a pair network based on characteristics of said pairs. It comprises calculating distances (di1) in a Euclidean space (E) between points (Mi) representative of these pairs.

Description

 Method and apparatus for interfacing with networks in a network

The invention lies in the general field of peer networks.

Thus, there is a large number of software programs that allow the user of a peer, in a peer network, to share files with users of other peers chosen from among his relatives or a circle of friends. In this document, we say that a peer "publishes" files when it allows files to be shared with other peers in the network.

However, no solution exists to date to connect users with each other without them knowing each other.

The present invention provides a solution that does not have this disadvantage by providing a method for connecting peers in a network without the users of these peers know each other, based on common interests such as common musical tastes .

More specifically, the invention relates to a method of relating a first peer in a peer network, with at least a second pair in this network. This process comprises:

a step in which the first pair is associated with a point representative of at least one characteristic of the first pair in a Euclidean space each dimension of which is defined from at least one type of characteristic; for each of a plurality of potential peers in the network:

a step during which this potential peer is associated with a representative point, in this Euclidean space, of characteristics of the potential peer; and

a step of calculating a distance in this Euclidean space between this point and the first point; and

a step of selecting the at least one second peer from the potential pairs as a function of these distances. This method can be implemented by a device such as a server. Accordingly, the invention also provides a device that can be used in a peer network to connect a first peer with at least a second pair of the network.

This device comprises: means for associating the first peer with a representative point, in a Euclidean space, with a plurality of characteristics of this first pair, each dimension of the space being defined at least from a type of feature ; means for associating each of a plurality of potential pairs in the network with a representative point, in this Euclidean space, of characteristics of the potential peer and for calculating a distance in this Euclidean space between this point and the first point; and

means for selecting the at least one second pair from among said potential pairs as a function of these distances. The method and the device according to the invention can thus select peers on the network and connect them by analyzing certain characteristics of these peers. These characteristics are peer-specific characteristics, as opposed to peer-dependent characteristics such as geographical distances or peer-to-peer network. Indeed the invention proposes the creation of a real map representing the peers in a space according to their own characteristics, for example the number of files of a type published in a network by each peer.

For example, the invention can be used to connect users according to the type of questions they ask on forums, depending on the type of software installed on their peers, traffic generated by these peers on the network or preferred hours of connection of these peers to the network.

In a particular mode of implementation of the invention, each dimension of the Euclidean space is associated with a file type, and the coordinate of the points in the Euclidean space according to this dimension is defined by the number of files of this file. type published in the network by the peer associated with that point.

In this document, the expression "type of file" is to be understood in a broad sense. It can in particular designate the types associated with extensions known to those skilled in the art, for example:

- type "image" for the extensions .jpg, .tif, .bmp; "

- "video" type for .mpg, .avi, ... extensions - "audio" type for .mp3 extensions, ...

In this embodiment, the invention thus relates people who choose to publish audio files rather than video files.

But, in the invention, the expression "type of file" can also designate different kinds of files having the same extension. For example, the invention makes it possible to distinguish, among the audio files .mp3, the music files of different genres, for example pop, rock, jazz.

The invention can thus connect jazz music fans without them knowing each other beforehand. In another embodiment of the invention, a coordinate of a point in Euclidean space is defined by the number of files of a type published in the peer network, divided by the total number of files published by that peer. peer.

This characteristic makes it possible to privilege the affinity of a user for a given type of file, for example music files of the jazz genre, rather than the quantity. Thus a user who has published a large amount of jazz music files can be put in touch with a user who publishes less but better matches his profile. In a particular embodiment, the peers selected to be related to a first pair are those that are associated with a point whose distance from the point associated with the first peer is less than a predetermined distance.

Thus, the invention makes it possible to connect all the users of peers having a common characteristic, without restriction on the number of these peers.

In another embodiment, to select the peers to be related to a first peer, a predetermined number of peers associated with the points closest to the point of that first peer are selected. In this embodiment, the invention makes it possible to relate a predetermined number of users.

In the detailed description of an embodiment that follows, we have chosen a three-dimensional Euclidean space, but the invention applies in a space having any number of dimensions.

In a particular embodiment, the different steps of the linking method are determined by instructions of computer programs. Consequently, the invention also aims at a computer program that can be implemented in a connection device or, more generally, in a server, this program comprising instructions adapted to the implementation of the steps of FIG. a method of relating as described above. This program can use any programming language, and be in the form of source code, object code, or intermediate code between source code and object code, such as in a partially compiled form, or in any other form desirable shape. The invention also relates to a computer-readable information medium, comprising instructions of a computer program as mentioned above.

The information carrier may be any entity or device capable of storing the program. For example, the medium may comprise storage means, such as a ROM, for example a PD

ROM or a microelectronic circuit ROM, or a magnetic recording means, for example a floppy disk or a hard disk.

On the other hand, the information medium may be a transmissible medium such as an electrical or optical signal, which may be conveyed via an electrical or optical cable, by radio or by other means. The program according to the invention can in particular be downloaded to a network of the Internet type.

Alternatively, the information carrier may be an integrated circuit in which the program is incorporated, the circuit being adapted to execute or to be used in the execution of the process in question.

Other features and advantages of the present invention will emerge from the description given below, with reference to the accompanying drawings which illustrate an embodiment having no limiting character. In the figures:

- Figure 1 shows, in its environment, a server according to the invention in a particular embodiment; FIG. 2 represents a Euclidean space used a particular mode of implementation of the invention; and

- Figure 3 shows, in flowchart form, the main steps of a connection method according to the invention in a particular embodiment.

Figu re 1 represents a server 10 according to the invention. This server 10 has the hardware architecture of a computer. It includes a processor 11, a random access memory 12, a read-only memory 13 and means 14 for communicating with peers Pi connected, such as this server 10, to a network of peers 100, which pairs can be materialized by communicating devices of various natures, such as personal computers, PDAs or mobile phones.

The read-only memory 13 constitutes a recording medium readable by the server 10 on which is recorded a computer program according to the invention, this program comprising instructions for carrying out the steps of a method of putting into proper relationship the invention, the main steps of which are represented in flowchart form in FIG. 3. In the embodiment described here, the invention makes it possible to relate peer users according to their musical tastes. Thus, FIG. 1 shows the number of music files published by the peer P ₁ in the network 100, namely:

- 28 music files of the pop genre; - 35 music files of the rock genre;

- 17 music files of the jazz genre. In the same way, the user of the peer P ₂ has chosen to publish on the network 100:

- 10 music files of the pop genre;

- 30 music files of the rock genre; - 5 music files of the jazz genre;

Finally, the peer user P ₃ publishes, on the network 10:

- 40 music files of the pop genre;

- 36 music files of the rock genre;

- 0 music files of the jazz genre. In FIG. 2, the points M ₁ , M ₂ and M ₃ associated with each of the pairs Pi, P ₂ and P _{3 are represented} in a Euclidean space E

In the embodiment described here, each dimension of the space E corresponds to a musical genre. As a result, space E has three dimensions, namely a pop dimension, a rock dimension and a jazz dimension.

The point M ₁ associated with the peer P ₁ therefore has coordinates {28, 35, 17} in this space.

It will be assumed that in this embodiment, it is sought to put the peer user P ₁ with at least one user of another peer of the network 100.

For this, we calculate the distance, in the Euclidean space E, between the point M ₁ associated with the peer P ₁ and each of the points M ₂ , M ₃ associated with the pairs P ₂ , P ₃ .

In the example described here, the distance d ,, _j between two points M ₁ , M _j of respective coordinates Jx ₁ , y ,, z,} and {X _j , V _j , Z _j } is calculated as follows :

d _U] = ^ ₁ - * f + (J ₁ - Jf + (Z ₁ - Z ₁ ) ² (1)

We thus find

We see in this example that the point M ₃ is closer to the point M ₁ than the point M ₂ . Also, if one chooses to put in relation with the user of the peer P ₁ the user having the musical tastes closest to it, one will select the user of the peer P ₃ .

We will now describe with reference to FIG 3 the main steps of the method of putting into relation according to the invention in a particular embodiment.

During a step E10, we define the dimensions of the Euclidean space E

In this example, each dimension of the Euclidean space E corresponds to a musical file type.

Alternatively, a dimension could correspond to a combination of music files for example.

During a step E20, we represent, in the Euclidean space E, the points associated with the different peers of the network. In the example described here, the coordinate of this point on each of the dimensions is equal to the number of music files of the kind associated with this dimension.

Then, during a step E30, one calculates the distances between all the points of the Euclidean space E taken two to two. Any distance in the mathematical sense of the term can be used.

We recall that strictly speaking a distance between two points M, N is a mathematical relation which satisfies:

D (M ₁ M) = 0; D (M ₁ N)>0; and

D (M ₁ P) <D (M ₁ N) + D (N ₁ P).

But in the context of this invention, the term distance must also be understood to encompass the result of a calculation of the type of equation (1), and wherein one or more weights have been introduced to favor one or more characteristics. For example :

d. ^ Λ ^ - x.Ϋ Hy ₁ - V,) ² Hz ₁ - Z ₁ Y Those skilled in the art will understand that in this example, a larger weight is given to the characteristic associated with the first coordinate X ₁ .

During a step E40, the peers are then selected to relate according to the distances between the points associated with these peers in the Euclidean space E

For example, the peers selected to be related to a peer P may be all peers whose points are contained in a sphere of predetermined radius centered on the point M associated with that peer.

The peers selected to be put in relation with the peer P ₁ may also be the peers whose points are closest to the point M ₁ . During a step E50, the selected peers are then connected.

For this purpose, in the embodiment described here, the server 10 sends an email to each of the peers of the network with the list of peers selected to be put in relation with this peer.

Claims

REVENDI CATI ONS A method of relating a first peer (P ₁ ) in a peer network (100), with at least a second peer (P ₂ ) in the network, the method comprising: a step during which said first pair (P ₁ ) is associated with a point (M ₁ ) representative of at least one characteristic of said first pair (P ₁ ), in a Euclidean space (E) of which each dimension (E ) is defined from at least one type of characteristic; for each of a plurality of potential pairs (P ₁ ) in said network: a step during which this potential pair (P ₁ ) is associated with a representative point (M ₁ ) in said Euclidean space (E) of characteristics of said potential pair (P ₁ ); and a step of calculating a distance (d, i) in said Euclidean space (E) between this point (M ₁ ) and said first point (M ₁ ); and a step of selecting said at least one second pair (P ₂ ) from said potential pairs (P ₁ ) as a function of said distances (d, -ι). 2. Method of linking according to claim 1 characterized in that each of said dimensions is associated with a type of files and in that the coordinate, according to this dimension, of a said point (M ₁ , M ₁ ) is defined by the number of said files of this type published in said network by said peer (P ₁ , P ₁ ) associated with this point (M ₁ , M ₁ ). 3. Method of linking according to claim 1, characterized in that each of said dimensions is associated with a file type and in that the coordinate, according to this dimension, of a said point (M ₁ , M ₁ ) is defined by the number of said files of this type published in said network by said peer (P ₁ , P ₁ ) associated with this point (M ₁ , M ₁ ), divided by the total number of files published by said peer (P ₁ , P ₁ ). 4. Method of putting into relation according to any one of claims 1 to 3, characterized in that said at least a second peer (P ₂ ) is selected if the distance (d ₂₁ ) between the point (M ₂ ) associated with this second pair (P ₂ ) and said first point (M ₁ ) is less than a predetermined distance. 5. Connecting method according to any one of claims 1 to 4, characterized in that during the selection step is performed a choice of a predetermined number of potential pairs (P ₁ ) associated with the points (M ₁ ) closest to said first point (M ₁ ). A computer program comprising instructions for performing the steps of the linking method according to any one of claims 1 to 5 when said program is executed by a computer (10). A computer-readable recording medium (13) (10) on which is recorded a computer program comprising instructions for executing the steps of the method of connection according to any one of claims 1 to 5. . 8. Device (10) capable of being used in a peer network (100) for connecting a first peer (P ₁ ) with at least a second peer (P ₂ ) of said network, said device (10) comprising: means (10) for associating said first peer (P ₁ ) with a representative point (M ₁ ) in a Euclidean space (E) with a plurality of characteristics of said first peer (P ₁ ), each dimension of said space (E) being defined at least from one type of characteristic; means for associating each of a plurality of potential pairs (P ₁ ) in said network with a representative point (Mi) in said Euclidean space (E) with characteristics of said potential pair (Pi) and for calculating a distance (d) , i), in the said Euclidean space (E) between this point (M ₁ ) and the said first point (M ₁ ); and means for selecting said at least one second pair (P ₂ ) from said potential pairs (P ₁ ) as a function of said distances (d, -ι).