CN106888159A - Message propagation path method for digging and system between the group based on online social platform group chat message - Google Patents

Abstract

The invention discloses a method and system for mining message propagation paths among groups based on group chat messages on an online social platform, and obtains group chat messages on an online social platform according to specified requirements through a data acquisition module, and hands over the obtained data to data processing module for processing; the data processing module saves the processed data in groups according to propagation events in the local storage module; the group chat message propagation path construction module extracts the required data from the local storage module to construct a message propagation path, and saves the result locally In the storage module; the visualization module extracts the group chat message propagation path from the local storage module or the message propagation path construction module, and visually displays it to the user in the form of a graph. The present invention makes up for the lack of the technology for mining propagation paths of specific messages between groups in the prior art.

Description

Mining method of message propagation path between groups based on group chat messages on online social platform with the system

technical field

The invention relates to a method and system for mining and constructing inter-group message propagation paths based on group chat messages on an online social platform, and belongs to the field of information technology.

Background technique

The development of online social software has enriched the way people communicate. Groups in online social software are generally created based on certain interests, purposes or social circles, while the group chat mode based on groups can provide a platform for multiple people to exchange information at the same time. The social attributes of human society lead to the fact that the relationship between groups is often not isolated. In some different groups created based on the same interest or purpose, some of the same group members are usually included at the same time. Different groups can establish connections through these common group members, and these common group members play a role of bridge in the process of message transmission between groups.

When information spreads within a group, the information may be forwarded to another group by a group member acting as a bridge, which causes the information to spread from one group to another. This behavior may occur multiple times, causing the information to spread from group to group. By mining and constructing the dissemination network of messages among groups, it is possible to trace the source of messages and find diffusion paths, etc. However, there is no relevant method to study the behavior of information dissemination among groups.

Contents of the invention

Purpose of the invention: In order to overcome the deficiencies in the prior art, the present invention provides a method and system for mining message propagation paths between groups based on group chat messages on an online social platform. This method and system can be based on the relationship between groups Topological structure, group chat message content, group chat message sender and message sending timing relationship and other characteristics, mining the propagation path of group chat messages among groups.

Technical scheme: in order to achieve the above object, the technical scheme adopted in the present invention is:

A method for mining message propagation paths between groups based on group chat messages on an online social platform, comprising the following steps:

Step 1: Select a specific group chat message as the propagation source of the propagation event;

Step 2: For each communication source, identify and extract group chat messages related to its content in the online social platform, and save them as a set of communication event messages;

Step 3: For each set of propagation event messages, according to the characteristics including the sending interval of group chat messages and whether the sender of group chat messages is a common group member between the two groups, determine the interval between each two group chat messages. Whether there is a communication relationship;

Step 4: Add a message propagation relationship between the groups to which each group chat message pair with a propagation relationship belongs, and construct and display a message propagation path.

Specifically, the group chat message includes at least features such as: group information, group chat message sender, group chat message sending time, group chat message sending time interval, and group chat message content.

Specifically, in the step 1, the specific group chat message is defined as: after preprocessing the specific content of the group chat message, including word segmentation, deleting stop words, and deleting words of a specific part of speech, the number of reserved words is greater than a certain Threshold group chat messages.

Specifically, in the step 1, the method of selecting the communication source includes, but is not limited to, the method of manually selecting a specific group chat message, the method of classifying different types of messages through a clustering algorithm, and the like.

Specifically, in the step 2, the set of propagation event messages is defined as: for a certain propagation source, the sending time of the group chat message is after the propagation source, and the content of the group chat message has a certain degree of similarity with the propagation source.

Specifically, in the step 2, the correlation between the group chat message and the communication source is confirmed by algorithms such as text similarity.

Specifically, in the step 4, the way of displaying the message propagation path includes a directed graph and the like.

A message propagation path mining system between groups based on group chat messages on an online social platform, including a data acquisition module, a data processing module, a local storage module, a message propagation path construction module, and a visualization module, wherein:

The data acquisition module is used to determine the communication source of the communication event, and for each communication source, search for group chat messages related to its content within a certain period of time in the online social platform;

The data processing module is used to process the group chat messages searched by the data acquisition module, including extracting the group information of the group chat messages, the sender of the group chat messages, the sending time of the group chat messages and the content of the group chat messages, and identifying the content related to the source of communication The group chat messages form a broadcast event message set and store it in the local storage module.

The local storage module is used to store various intermediate data and final result data;

The message propagation path construction module uses the method for mining message propagation paths between groups based on group chat messages on an online social platform to construct a message propagation path.

The visualization module visually displays the message propagation path.

The data acquisition module obtains the group chat message of the online social platform according to the specified requirements, and hands over the obtained data to the data processing module for processing; the data processing module stores the processed data in the local storage module according to the propagation event grouping; The chat message propagation path construction module extracts the required data from the local storage module to construct the message propagation path, and saves the result in the local storage module; the visualization module extracts the group chat message propagation path from the local storage module or the message propagation path construction module, And visually display it to the user in the form of a graph.

Beneficial effects: The method and system for mining message propagation paths between groups based on group chat messages on online social platforms provided by the present invention can mine out the message propagation path by analyzing the content, sender, sending time, and group messages of group chat messages and other characteristics of group chat messages. The propagation path between groups is finally displayed through the visualization module; the present invention makes up for the lack of technology in the prior art for mining the propagation path of group chat messages between groups, and proposes a method that is simple to implement, A solution that works well.

Description of drawings

Fig. 1 is the flow chart of main body method of the present invention;

Fig. 2 is a schematic structural diagram of the system of the present invention;

Fig. 3 is the specific step flowchart of step 1 in the subject method of the present invention;

Fig. 4 is the specific step flowchart of step 2 in the subject method of the present invention;

Fig. 5 is the specific step flowchart of step 3 in the subject method of the present invention;

Fig. 6 is a schematic diagram of the content of all group chat messages in a broadcast event message set;

Fig. 7 is a schematic diagram of the relationship between the source message to be analyzed and the current target message set to be analyzed;

FIG. 8 is a schematic diagram of a message propagation path network.

detailed description

The present invention will be further described below in conjunction with the accompanying drawings.

As shown in Figure 1, it is a method for mining message propagation paths between groups based on group chat messages on online social platforms, including the following steps:

Step S1: Select a specific group chat message as the source of the propagation event; as shown in Figure 3, it specifically includes the following steps:

Step S1.1: Determine whether to select the propagation source of the propagation event manually or by a clustering algorithm: if it is manually selected, go to step S1.6; otherwise, go to step S1.2;

Step S1.2: The user sets parameters such as group set and time span, and proceeds to step S1.3;

Step S1.3: Use the clustering algorithm to cluster the group chat messages of the selected group within the selected time period, and sort the group chat messages in each clustering result set in chronological order, and proceed to step S1.4 ;

Step S1.4: Perform word segmentation and part-of-speech tagging on the group chat messages in the clustering result set, delete stop words, and extract the part-of-speech words such as person names, place names, organization names, and proper nouns that appear in the communication source to form the group Keyword collection of chat messages, enter step S1.5;

Step S1.5: For each clustering result set, select the group chat message with the number of keywords greater than a certain threshold and the earliest sending time as the propagation source of the result set, and proceed to step S1.6;

Step S1.6: Determine and save the manually selected propagation source of the propagation event or the propagation source of the propagation event calculated by the clustering algorithm.

Step S2: For each communication source, identify and extract group chat messages related to its content in the online social platform, and save them as a set of communication event messages; as shown in Figure 4, specifically include the following steps:

Step S2.1: The user sets the group set to be analyzed and the time span parameters, and proceeds to step S2.2;

Step S2.2: Select a propagation source of the propagation event, take out the group chat messages within the selected group and time span one by one, calculate the text similarity between the propagation source of the propagation event and each group chat message, and proceed to step S2.3;

Step S2.3: Save group chat messages whose text similarity is greater than a certain threshold in the propagation event message set, and proceed to step S2.4;

Step S2.4: Judging whether all propagation sources of propagation events have been processed: if not, go to step S2.2; otherwise, go to step S2.5;

Step S2.5: sort and save the group chat messages in each broadcast event message set according to the order of message sending time.

Step 3: For each set of propagation event messages, according to the characteristics including the sending interval of group chat messages and whether the sender of group chat messages is a common group member between the two groups, determine the interval between each two group chat messages. Whether there is a communication relationship; as shown in Figure 5, it specifically includes the following steps:

Step S3.1: Extract all group chat messages in a broadcast event message set; Figure 6 shows a schematic diagram of group chat messages in a certain broadcast event message set extracted, including group chat message sending time (M401), group chat message Data such as message sender (M402), sender's group information (M403) and group chat message content (M404), the group chat messages in the dissemination event message set are sorted according to the order of sending time;

Step S3.2: Traversing the set of propagation event messages, first select a group chat message (called "source message to be analyzed" (M405)) according to the order of group chat message sending time, and then extract the sending time after the group chat message All group chat messages within a period of time (referred to as "target message collection to be analyzed" (M406);

Step S3.3: As shown in Figure 7, traverse the target message set to be analyzed, select a group chat message in the set (called "current target message to be analyzed" (M408)), and determine whether the message is related to the source to be analyzed Whether the message sending time difference of the message is less than the threshold, and at the same time judge whether the sender of the message is also a group member of the group to which the source message to be analyzed belongs;

Step S3.4: Save the source message to be analyzed and the current target message to be analyzed that satisfy step S3.3 as a pair of message pairs with a propagation relationship, and the propagation direction is from the source message to be analyzed to the current target message to be analyzed;

Step S3.5: Determine whether the source message to be analyzed is the last message in the dissemination event message set: if not, go to step S3.2; otherwise, go to step S3.6;

Step S3.6: output message pairs with propagation relationship.

Step 4: Add a message propagation relationship between the groups to which each group chat message pair with a propagation relationship belongs, construct and display a message propagation path; the constructed message propagation path is shown in Figure 8, and the group in this step The construction and representation of propagation relationships include but not limited to directed graphs and other forms; circles in the figure represent groups (M501, M503), and directed edge tables (M502) represent message propagation directions.

The cooperation relationship between the modules of the present invention will be described in detail below with reference to the accompanying drawings.

FIG. 2 shows a block diagram of a system for mining message propagation paths between groups based on group chat messages on an online social platform according to the present invention. This system is composed of data acquisition module M1, data processing module M2, local storage module M3, message transmission path construction module M4, visualization module M5, etc. The interrelationships between the modules are as follows.

Data acquisition module M1: This module can determine the source of propagation events through manual selection or machine cluster analysis according to user settings; for each source of propagation events, search For group chat messages in social platforms, calculate and save group chat messages similar to those between the propagation sources of propagation events.

Data processing module M2: used to process the group chat messages obtained by the data acquisition module; the processing includes extracting the information of the group to which the group chat messages belong, the sender of the group chat messages, the sending time, and the sending content, and at the same time identifying the information similar to the source message For the group chat messages of the content, the group chat messages are grouped according to the propagation events according to the timing relationship, and stored in the local storage module.

The local storage module M3 is used to store various intermediate data and final result data.

Message propagation path construction module M4, for using the message propagation path mining method between groups based on online social platform group chat messages of the present invention to construct message propagation paths between groups, and store the obtained results in the local storage module , or directly to the visualization module.

The visualization module M5 extracts the constructed message propagation path data from the local storage module or the message propagation path construction module, and visualizes them in the form of charts and the like.

The above is only a preferred embodiment of the present invention, it should be pointed out that for those of ordinary skill in the art, without departing from the principle of the present invention, some improvements and modifications can also be made, and these improvements and modifications are also possible. It should be regarded as the protection scope of the present invention.

Claims (8)

1. message propagation path method for digging between a kind of group based on online social platform group chat message, it is characterised in that：Bag Include following steps：

Step 1：Specific group chat message is chosen as communication eventses propagating source；

Step 2：To each propagating source, related to its content group chat message in online social platform is recognized and extracts, and protect Save as a communication eventses massage set；

Step 3：To each communication eventses massage set, sent according to including group chat message transmission time interval, group chat message Person whether be common group members between two groups in interior feature, determine whether to have to propagate between each two group chat message to close System；

Step 4：Relation is propagated to adding message between affiliated group in each group chat message with propagation relation, is built and is shown Message propagation path.

2. message propagation path excavation side between the group based on online social platform group chat message according to claim 1 Method, it is characterised in that：The group chat message at least includes such as feature：Group information, group chat sender of the message, group chat message send Time, group chat message transmission time interval, group chat message content.

3. message propagation path excavation side between the group based on online social platform group chat message according to claim 1 Method, it is characterised in that：In the step 1, specific group chat message definition is：The particular content of group chat message include point Word, delete stop words, delete the word of specific part of speech after interior pretreatment, the group of the word quantity more than certain threshold value of reservation Merely message.

4. message propagation path excavation side between the group based on online social platform group chat message according to claim 1 Method, it is characterised in that：In the step 1, the choosing method of propagating source is including manually choosing specific group chat message approach, by poly- Class algorithm partition different types of messages method.

5. message propagation path excavation side between the group based on online social platform group chat message according to claim 1 Method, it is characterised in that：In the step 2, communication eventses massage set is defined as：To certain propagating source, when group chat message sends Between after propagating source and group chat message content and the propagating source have the group chat massage set of certain similarity.

6. message propagation path excavation side between the group based on online social platform group chat message according to claim 1 Method, it is characterised in that：In the step 2, group chat message is confirmed with the correlation of propagating source by text similarity algorithm.

7. message propagation path excavation side between the group based on online social platform group chat message according to claim 1 Method, it is characterised in that：In the step 4, the mode for showing message propagation path is directed graph.

8. message propagation path digging system between a kind of group based on online social platform group chat message, it is characterised in that：Bag Include data acquisition module, data processing module, module, message propagation path structure module and visualization model is locally stored, its In：

Data acquisition module, for determining communication eventses propagating source, and to each propagating source, searches for certain in online social platform The group chat message related to its content in the individual time period；

Data processing module, for the group chat message that processing data acquisition module is searched, including the group for extracting group chat message Information, group chat sender of the message, group chat message send time and group chat message content, the identification group chat related to source contents are propagated Message, forms the merging of communication eventses message set and stores in module is locally stored.

Module is locally stored, for storing various intermediate data and final result data；

Message propagation path builds module, using message propagation path between the group based on online social platform group chat message Method for digging builds message propagation path.

Visualization model, visual presentation message propagation path.