Realtime incremental detection method and system for social network events
A social network and detection method technology, applied in the field of real-time incremental detection of social network events, can solve the problems of not being able to adapt to the real-time nature of short text, socialization and fragmentation characteristics, and inaccurate detection results at the same time
Active Publication Date: 2015-01-14
BEIHANG UNIV
View PDF1 Cites 1 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0007] The present invention provides a real-time incremental detection method and system for social network events, which are used to solve the problem that the event detection in the prior art cannot simultaneously adapt to the real-time, socialization and fragmentation characteristics of short texts in social networks, resulting in Technical problems with inaccurate test results
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
[0023] figure 1 A schematic flow diagram of a real-time incremental detection method for social network events provided by an embodiment of the present invention, as shown in figure 1 shown, including:
[0024] 101. Using a probabilistic graph model, according to the time, document and topic tags of the short text, model learning is performed on the short text to obtain a likelihood function.
[0025] 102. Using an Expectation Maximization Algorithm (EM) algorithm to solve the likelihood function to obtain parameters.
[0026] Among them, the parameters include p(z|d), p(t d |z), p(h|z) and p(w|z), and p(z|w, d, t d , h) and p(z|w, d, t d ); among them, p(z|d) represents the probability of topic z in document d; p(t d |z) means topic z at time t d , p(h|z) represents the probability that the topic label h appears in the topic z, and p(w|z) represents the probability of the word w in the topic z. p(z|w, d, t d , h) means topic z involves word w, document d, time t 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 invention provides a realtime incremental detection method and system for social network events. The method includes performing model learning on a short text according to time, a document and a theme label of the short text by adopting a probability graph model to acquire a likelihood function; adopting an EM algorithm to solve the likelihood function to acquire parameters; adopting an increment updating mode to perform iterative updating on the acquired parameters until the parameters are converged; adopting a distributed-type mode to execute a step E and a step M in the EM algorithm according to the parameters after convergence, and calculating to acquire content of the short text. The technical problem of inaccuracy in detection result due to the fact that event detection cannot adapt to features of instantaneity, socialization and fragmentation of the short text in a social network in the prior art is solved. A supervised short text event detection model, an algorithm combing increment learning with prediction and an event detection model based on an internal memory calculation platform are provided.
Description
technical field [0001] The invention relates to information technology, in particular to a real-time incremental detection method and system for social network events. Background technique [0002] Short texts in social networks, such as Weibo, often have the following characteristics: the length is strictly limited within 140 characters; users can interact with other users through symbols while posting short texts; users can also use the # symbol to indicate short texts The topic this book belongs to. [0003] As a highly interactive and communicative tool, the number of short texts in social networks often shows explosive growth with the occurrence of news events, which makes real-time information on social networks more frequent; at the same time, due to social The length limit of short text in the network makes the text more fragmented. In general, the real-time, social and fragmented characteristics of short texts in social networks have brought great challenges to ev...
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 Applications(China)
IPC IPC(8): G06F17/30
CPCG06Q50/01
Inventor 李建欣邰振赢于伟仁张日崇胡春明
Owner BEIHANG UNIV