CN102855272B - A kind of micro-blog contains the D-S evidence theory method of traffic information fusion - Google Patents

Abstract

The invention discloses a kind of micro-blog and contain the D S Method of Evidence Theory of traffic information fusion, described method includes: according to congested in traffic degree, conclusion evidence theory framework of identification and proposition space；According to time window and road section, capturing the micro-blog content relevant to transport information theme, composition needs the micro-blog data set merged；Micro-blog data set pretreatment；Calculate micro-blog data set traffic behavior and describe the acceptation similarity of vocabulary, introduce Chinese corpus resource, improve acceptation similarity computational accuracy；Calculate micro-blog acceptation similarity weighting evaluation, build evidence theory Basic probability assignment function；According to Dempster rule of combination to a plurality of micro-blog evidence source, carrying out combining evidences, the trust of each proposition in determining identification space is interval, chooses and trusts proposition that in interval, belief function is maximum as fusion results.Utilizing the present invention can realize micro-blog and contain the fusion of transport information, the collection for urban traffic information provides a kind of significant data source.

Description

A kind of micro-blog contains the D-S evidence theory method of traffic information fusion

Technical field

The present invention relates to mobile location-based service, the Internet space information search, mobile internet technology, be specifically related to a kind of micro-blog and contain the D-S evidence theory method of traffic information fusion.

Background technology

Real-time Traffic Information can be alleviated traffic congestion, improve traffic transportation efficiency, ensure traffic safety, facilitate Public Traveling, existing transport information obtains mode and mainly includes fixing sensor technology (induction coil, video monitoring and microwave sounding), GPS and the floating car technology of wireless telecommunications system, communication terminal signalling analysis technology etc. it is installed, but these acquisition means are obtaining temporary traffic control restricted information and tackling the aspects such as sudden traffic events and still suffer from the biggest limitation.Micro-blog contains ageing the highest abundant Real-time Traffic Information, contain various transport information type, such as include that road traffic flow, the coast is clear degree and travel speed, traffic restriction, temporary traffic control, sudden traffic events, traffic behavior for locality describe information etc., obtain the high dynamic Real-time Traffic Information contained in micro-blog and can make up the deficiency of existing traffic information collection means.

But, the description diversity that the high dynamic of micro-blog message, ambiguity and different micro-blog user thereof give out information makes information fusion become the bottleneck problem of information retrieval, directly affects micro-blog and contains the utilization of Real-time Traffic Information.Micro-blog traffic information fusion is that the transport information content being contained different micro-blog message makes inferences decision-making, obtains traffic behavior accurately and describes information, preferably serves traffic administration and trip service.The difficult point of micro-blog traffic information fusion is: the destructuring feature of (1) micro-blog message causes semantic understanding difficulty: owing to micro-blog message content is simplified, about only 140 words, and colloquial style feature is obvious, containing more redundant content, pose a big pressure with extraction to the Semantic judgement of automatization；(2) the different micro-blog users description difference to traffic behavior that gives out information causes information taken contradiction: in certain period of time, it is understood that there may be describe a plurality of micro-blog message of same road section traffic volume state.For same road conditions, the description of different user may be widely different, and some describes even semanteme and repels each other.

For solving this problem, the technology used at present is Text Clustering Method, and when text has certain vocabulary quantity, text cluster process can determine that text subject describes accurately.But micro-blog message content is short and small, after the processes such as participle, word sense disambiguation, meaning of a word mummification, it is possible to use traffic behavior to describe key vocabularies little.Therefore, text cluster can not solve ambiguity and the diversity of different micro-blog user message description of micro-blog message semantic definition well.

To this end, this patent is for above-mentioned information fusion problem, a kind of micro-blog traffic information fusion method based on D-S evidence theory is proposed.The method, by introducing Chinese corpus resource, enriches the semantic information of micro-blog, solves the fuzzy problem that micro-blog describes.On Chinese corpus Knowledge Base, achieve and micro-blog content is carried out acceptation similarity weighting evaluation, the uncertain inference problem of the information fusion caused due to the diversity of different microblog users is processed followed by evidence theory, so that it is determined that fusion results.Collect to dynamic information and provide a kind of new solution, make up the link that traditional dynamic traffic information collecting technology is relatively weak.

Summary of the invention

The technical problem to be solved in the present invention is: be difficult to fully merge utilization, conventional dynamic transport information collection method for a large amount of transport information contained in current micro-blog, is difficult to react in time the present situation of paroxysmal traffic information.The present invention proposes a kind of micro-blog and contains the D-S evidence theory method of traffic information fusion, provides another important data source for collecting of dynamic information.This method solving the destructuring feature of micro-blog message causes semantic understanding difficult and the different micro-blog user description difference to traffic behavior that gives out information causes information taken contradiction, may be directly applied to individual and vehicle mounted guidance, mobile location-based service, map web site, trip information service platform, Logistic Scheduling and the emergency traffic prediction scheme of specialty.

The technical solution of the present invention is:

A kind of micro-blog contains the D-S evidence theory method of traffic information fusion, including:

According to congested in traffic degree, conclusion evidence theory framework of identification Θ and proposition space 2^Θ；

According to effective time window T_intervalWith road network section road, capturing the micro-blog content relevant to transport information theme, composition needs the micro-blog data set V merged；

Micro-blog information V_iPretreatment operation, including natural language participle, word sense disambiguation, meaning of a word mummification, obtains micro-blog traffic behavior descriptor and collects W_i；

Introduce Chinese corpus resource Corpus={C_wikipedia, C_hownet..., C_num, calculate micro-blog traffic behavior word finder W_iWith the acceptation similarity Sim of vocabulary in proposition space；

Calculate micro-blog message V_iAcceptation similarity weighting evaluation Score_i, conclusion evidence theory Basic probability assignment function m (V_i)；

Carry out combining evidences and evidence decision by Dempster compositional rule, determine that this section road micro-blog contains traffic information fusion result TState_road；

Preferably, described include according to congested in traffic degree conclusion evidence theory framework of identification:

The determination of congested in traffic degree can refer to Countries standard, and road network traffic congestion degree is classified by " the urban traffic control assessment indicator system " within 2002, announced such as the Ministry of Public Security；

The determination of congested in traffic degree is relevant to actual fused demand；

Evidence theory framework of identification and proposition space are not intended to size；

Preferably, described according to effective time window and road network section, capture transport information theme relevant microblog visitor, composition needs to merge micro-blog data set V and includes:

Effective time window T is defined as transport information time T_currentThe time period carrying out expanding and formed, i.e. T=[T_current-Δt_a, T_current+Δt_b] wherein Δ t_aWith Δ t_bUser's defined parameters；

In road network, section is with road network road name as object of study or with the roadway segment in navigation road network as object of study；

Micro-blog data set V is relevant to the actual road network section chosen；

The building process of micro-blog data set V is that this patent is not related to；

Micro-blog data set V is not intended to its storage form, can be data base or data file；

Preferably, described to introducing Chinese corpus resource Corpus, calculate micro-blog traffic behavior word finder Wi and include with the acceptation similarity Sim of vocabulary in proposition space:

Chinese corpus resource Corpus is not intended to corpus type, can be wikipedia, know net etc.；

The calculating process of acceptation similarity Sim is that this patent is not related to；

Preferably, the acceptation similarity Similarity-Weighted of described micro-blog content evaluates Score, determines the Basic probability assignment function m (V of micro-blog_i) including:

Micro-blog content meaning of a word Similarity-Weighted evaluation calculation, particularly as follows:

${\mathrm{Score}}_k=\underset{i=1}{\overset{T}{\mathrm{\Σ}}}\mathrm{Sim}({\mathrm{term}}_i,k)\·\log \left(\frac{\mathrm{num}\left(k\right)}{\mathrm{sum}}\right)\·\mathrm{boost}\left(\mathrm{user}\right)$

Wherein, k is the proposition in evidence policy innovation space, and term micro-blog traffic behavior descriptor collects W_iVocabulary, sum is the quantity of micro-blog, and num (k) is the micro-blog quantity comprising proposition k, boost (user) is the excitation function of this micro-blog user, reflecting the significance level of this user, default value is 1, and this user of the biggest explanation is the most important for this value.

The Basic probability assignment function of micro-blog content calculates, particularly as follows:

$m\left(V_i\right)=\frac{{\mathrm{Score}}_k}{\underset{k=1}{\overset{K}{\mathrm{\Σ}}}{\mathrm{Score}}_k}$

Preferably, described method also includes:

The section to be merged provided according to user and time of fusion condition, complete described micro-blog and contain traffic information fusion；

According to section and time of fusion condition in the wanted integration region that user provides, complete described micro-blog and contain traffic information fusion；

Present invention advantage compared with prior art is: instant invention overcomes micro-blog and contains the shortcoming that transport information is difficult to fully merge utilization；By introducing Chinese corpus resource, abundant micro-blog semantic understanding process, the destructuring feature efficiently solving micro-blog message causes semantic understanding difficulty；Utilize D-S evidence theory advantage in terms of processing the expression of uncertain information and synthesis, solve the different micro-blog user description difference to traffic behavior that gives out information and cause information taken contradiction.

Accompanying drawing explanation

Fig. 1 is that embodiment of the present invention micro-blog contains traffic information fusion D-S evidence theory method flow diagram；

Fig. 2 is embodiment of the present invention case embodiment flow chart

Fig. 3 is embodiment of the present invention case micro-blog information record set

Fig. 4 is embodiment of the present invention case wikipedia data source

Detailed description of the invention

In order to make those skilled in the art be more fully understood that the scheme of the embodiment of the present invention, with embodiment, the embodiment of the present invention is described in further detail below in conjunction with the accompanying drawings.

As it is shown in figure 1, be a kind of micro-blog of embodiment of the present invention D-S evidence theory method flow diagram of containing traffic information fusion, comprise the following steps:

Step 101, according to congested in traffic degree, conclusion evidence theory framework of identification Θ and proposition space 2^Θ；

Step 102, according to effective time window T_intervalWith road network section road, capturing the micro-blog content relevant to transport information theme, composition needs the micro-blog data set V merged；

Step 103, micro-blog information V_iPretreatment operation, including natural language participle, word sense disambiguation, meaning of a word mummification, obtains micro-blog traffic behavior descriptor and collects W_i；

Step 104, introduces Chinese corpus resource Corpus={C_wikipedia, C_hownet..., C_num, calculate micro-blog traffic behavior word finder W_iWith the acceptation similarity Sim of vocabulary in proposition space；

Step 105, calculates micro-blog message V_iAcceptation similarity weighting evaluation Score_i, conclusion evidence theory Basic probability assignment function m (V_i)；

Step 106, carries out combining evidences and evidence decision by Dempster compositional rule, determines that this section road micro-blog contains traffic information fusion result TState_road；

The actual application of the embodiment of the present invention is described further below in detail with the citing of Beijing's road network.

As in figure 2 it is shown, be a kind of micro-blog of present invention D-S evidence theory embodiment of the method case embodiment flow chart of containing traffic information fusion.The present embodiment is implemented under premised on present embodiment, gives detailed embodiment and concrete operating process, but protection scope of the present invention is not limited to following embodiment, and the detailed process of the present embodiment comprises the following steps:

Step 201, use for reference regulation in " urban traffic control evaluation is mark system " that the Ministry of Public Security of China announces for 2002, its congested in traffic degree mainly comprises following four: unimpeded (motor vehicles average overall travel speed is more than 30km/h), slight crowding (motor vehicles average overall travel speed 20～30km/h), crowded (motor vehicles average overall travel speed 10～20km/h), block up (motor vehicles average overall travel speed is less than 10km/h).Owing to micro-blog content is that the overview to traffic describes, not being difficult to be accurate to speed step, in order to ensure the precision of micro-blog information fusion, merged with crowded by slight crowding, definition framework of identification is；

Θ=unimpeded, crowded, block up

Then proposition space 2^ΘFor:

Step 202, all sections in searching loop Beijing road network, East 5th Ring Road section in chosen place line of reasoning net；

Step 203, it is judged that in road network, all sections have been disposed, if it is, perform step 217；If it is not, then perform step 204；

Step 204, build time window, set up and the micro-blog information record set with East 5th Ring Road traffic relevant information theme, as shown in Figure 3 about the micro-blog content of Beijing's dynamic information by search Sina micro-blog, Netease's micro-blog, Sohu's micro-blog and Tengxun's micro-blog etc.；

Step 205, micro-blog information pre-processing, natural language participle, word sense disambiguation, meaning of a word mummification, obtain micro-blog traffic behavior and describe vocabulary, then perform step 206；

Step 206, download and resolve wikipedia data source (as shown in Figure 4), use document (Milne D, Witten IH.An effective low-cost measure of semantic relatedness obtained from Wikipedia links.In Proceedings of the AAAI 2008 Workshop on Wikipedia and Artificial Intelligence) method set up semantic model, calculate the semantic similarity of micro-blog traffic behavior vocabulary, then perform step 207；

Step 207, calculates micro-blog message acceptation similarity weighting evaluation, the Basic probability assignment function that conclusion evidence is theoretical, then perform step 208；

Step 208, carries out combining evidences and evidence decision by Dempster compositional rule, determines that this section micro-blog contains traffic information fusion result, then performs step 209；

Step 209, it is judged that all sections, Beijing have been disposed, if it is, perform step 217；If it is not, then perform step 210；

Step 210, terminates this calculating process；

Visible, a kind of micro-blog of the embodiment of the present invention contains the D-S evidence theory method of traffic information fusion, micro-blog can be solved and contain that the semantic understanding during traffic information fusion is difficult and description variability issues between many microblog users, a kind of new information gathering scheme that provides, accurate fast reaction burst traffic information are provided for dynamic information.Present invention can be directly applicable to the issue of dynamic information, serve map web site system, public trip information platform and mobile location-based service.

It should be noted that the method for the embodiment of the present invention is applicable to the Dynamic Information Gathering of all urban road networks；The present invention is not intended to the Grasp Modes of micro-blog content and the offer website of micro-blog content；The present invention is not limited solely to the acceptation similarity weighting evaluation model of the acceptation similarity calculating method employed in detailed description of the invention and microblog users.

One of ordinary skill in the art will appreciate that all or part of step realizing in above-described embodiment method can be by program and completes to instruct relevant hardware, described program can be stored in a computer read/write memory medium, described storage medium, such as: ROM/RAM, magnetic disc, CD etc..

Being described in detail the embodiment of the present invention above, the present invention is set forth by detailed description of the invention used herein, the method that the explanation of above example is only intended to help to understand the present invention；Simultaneously for one of ordinary skill in the art, according to the thought of the present invention, the most all will change, in sum, this specification content should not be construed as limitation of the present invention.

Claims (6)

1. a micro-blog contains the D-S evidence theory method of traffic information fusion, it is characterised in that step includes:

According to congested in traffic degree, conclusion evidence theory framework of identification Θ and proposition space 2^Θ；

According to effective time window T_intervalWith road network section road, capturing the micro-blog content relevant to transport information theme, composition needs the micro-blog data set V merged；

Micro-blog message V_iPretreatment operation, including natural language participle, word sense disambiguation, meaning of a word mummification, obtains micro-blog traffic behavior word finder W_i；

Introduce Chinese corpus resource Corpus={C_wikipedia, C_hownet... }, calculate micro-blog traffic behavior word finder W_iWith the acceptation similarity Sim of vocabulary in proposition space；

Calculate micro-blog message V_iAcceptation similarity weighting evaluation Score_i, conclusion evidence theory Basic probability assignment function m (V_i)；

Carry out combining evidences and evidence decision by Dempster compositional rule, determine that this section road micro-blog contains traffic information fusion result TState_road。

Method the most according to claim 1, it is characterised in that described include according to congested in traffic degree conclusion evidence theory framework of identification:

Road network traffic congestion degree is classified by " urban traffic control assessment indicator system " that the determination of congested in traffic degree is announced with reference to the Ministry of Public Security for 2002；

The determination of congested in traffic degree is relevant to actual fused demand；

Evidence theory framework of identification and proposition space are not intended to size.

Method the most according to claim 1, it is characterised in that described according to effective time window and road network section, captures the micro-blog content relevant to transport information theme, and composition needs to merge micro-blog data set V and includes:

Effective time window T_intervalIt is defined as transport information time T_currentThe time period carrying out expanding and formed, i.e. T_interval=[T_current-Δt_a, T_current+Δt_b] wherein Δ t_aWith Δ t_bFor user's defined parameters；

In road network, section is with road network road name as object of study or with the roadway segment in navigation road network as object of study；

Micro-blog data set V is relevant to the actual road network section chosen；

Micro-blog data set V is not intended to its storage form.

Method the most according to claim 1, it is characterised in that described to introducing Chinese corpus resource Corpus, calculates micro-blog traffic behavior word finder W_iInclude with the acceptation similarity Sim of vocabulary in proposition space:

Chinese corpus resource Corpus is not intended to corpus type.

Method the most according to claim 1, it is characterised in that the acceptation similarity weighting evaluation Score of described micro-blog content, determines the Basic probability assignment function m (V of micro-blog_i) including:

Micro-blog content meaning of a word Similarity-Weighted evaluation calculation, particularly as follows:

Wherein, k is the proposition in evidence policy innovation space, and term micro-blog traffic behavior descriptor collects W_iVocabulary, sum is the quantity of micro-blog, and num (k) is the micro-blog quantity comprising proposition k, boost (user) is the excitation function of this micro-blog user, reflecting the significance level of this user, default value is 1, and this user of the biggest explanation is the most important for this value；

The Basic probability assignment function of micro-blog content calculates, particularly as follows:

6. according to the method described in any one of claim 1 to 5, it is characterised in that described method also includes:

The section to be merged provided according to user and time of fusion condition, complete described micro-blog and contain traffic information fusion；

According to section and time of fusion condition in the wanted integration region that user provides, complete described micro-blog and contain traffic information fusion.