A method and system for discovering causal relationship between social network users combining behavior sequence and text information
A text information and causal relationship technology, applied in unstructured text data retrieval, network data retrieval, text database query, etc., can solve the problems of misleading information transmission and data ineffectiveness, so as to reduce misleading and inaccurate results. , the effect of reducing false results
Active Publication Date: 2021-08-27
GUANGDONG UNIV OF TECH
View PDF3 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
In addition, in the use of time series data, the existing technology basically divides the data into equal intervals. Due to the sparsity, some data will have little effect or even mislead the discovery of information transmission. Therefore, we can use a more appropriate method to find the optimal interval sequence, reconstructing the data
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0044] The specific embodiment of the present invention will be further described below in conjunction with accompanying drawing:
[0045] Such as figure 1 As shown, a method for discovering causal relationship among social network users that combines behavior sequence and text information includes the following steps:
[0046] S1), write a web crawler through the framework of python+scrapy, construct the target URL according to the user ID, grab the user information and the dynamic data released in the target social network; this implementation mode takes Sina Weibo as an example, by analyzing Sina Weibo The URL of the user’s Weibo page and its parameters are constructed, and the URL corresponding to the user is constructed to start capturing data; the original Weibo data is cleaned, and some data with incomplete information is removed. The cleaned data has two dimensions, namely user and time;
[0047] S2), according to the time information of the data, the data is first d...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The present invention provides a method and system for discovering causal relationships between social network users by combining behavior sequences and text information, including: S1), data acquisition; S2), preprocessing the data at equal intervals in the smallest time unit; S3) , utilize the time series behavior data, optimize the objective function to find the optimal interval; S4) reconstruct the text data by splicing the text at the time of merging, and the text is represented by vectorization; S5) carry out the transfer entropy calculation to the text vector sequence of two or two users; S6 ), pruning to obtain the user causal relationship network; S7), user causal network storage and export; S8) user causal relationship query and visualization. The invention solves the problem caused by the sparse user activities to transfer entropy calculation; infers the user causal relationship of the social network with text data, and the amount of information is more abundant than pure behavior data; provides an interactive user causal relationship inference, query and export system.
Description
technical field [0001] The invention relates to the technical field of data mining, in particular to a method and system for discovering causal relationships between social network users combined with behavior sequences and text information. Background technique [0002] On social networks, due to the participation of a large number of users, a series of user activity data and user text, audio, and video data are generated. Users can post messages or upload pictures and videos through various online channels. Users can write what they see, hear, and feel in a sentence, and share it with friends anytime, anywhere through a computer or mobile phone; they can also follow their friends' dynamics. [0003] As the number of social network users continues to increase, the user's hobbies and topics of concern can be mined from the user's dynamics. People in data mining and analysis are also increasingly focusing on social networks. There are many scholars at home and abroad who st...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06F16/33G06F16/955
Inventor 蔡瑞初谢泳陈薇郝志峰陈炳丰
Owner GUANGDONG UNIV OF TECH