CN103092918A - Construction and query method of knowledge base of emergency auxiliary decision-making system based on location information - Google Patents

Abstract

The invention relates to the technical field of assistant decision-making, in particular to a method for constructing and inquiring an emergency assistant decision-making system knowledge base based on position information. The invention firstly extracts the knowledge or network knowledge in the historical knowledge base which forms the knowledge source of the position information knowledge base, and extracts the address information in the knowledge or network knowledge; then analyzing the address information, and finally determining heterogeneous effective address information; carrying out knowledge classification on the information analysis result according to address coding specifications, assigning longitude and latitude information according to the address coding specifications, and establishing an incidence relation table of knowledge and position information in a database; thereby forming a knowledge base based on the location information. The emergency assistant decision-making system is provided by utilizing the characteristic that the emergency has regionality and the correlation between knowledge and position information, so that the problems that the information is classified and displayed according to the position correlation and the like are solved; the method can be applied to an emergency assistant decision system.

Description

Construction and query method of knowledge base of emergency auxiliary decision-making system based on location information

technical field

The invention relates to the technical field of auxiliary decision-making, in particular to a method for building and querying a knowledge base of an emergency auxiliary decision-making system based on location information.

Background technique

With the development of social economy, the complexity of emergencies also arises. One of its manifestations is the frequent occurrence and diversification of public emergencies. Natural disasters, accident disasters, public health incidents, social security incidents and other emergencies in real life not only pose a huge threat to normal human activities, but also seriously affect the normal operation of the entire social and economic system. The timeliness requirements for emergencies are very high, and the processing involves a large amount of data. An emergency decision-making system is needed to efficiently analyze the data of emergencies and quickly form an optimal disposal plan.

Emergency auxiliary decision-making mainly goes through three stages of development: 1. Decision makers rely on their subjective ability, experience, and intuition as the basis for all decision-making analysis, and comprehensively use objective laws of the cognitive process such as logical reasoning to guide the decision-making process; 2. By introducing some The method in the field of mathematics and systems engineering, according to the characteristics of different emergencies, takes into account the environmental factors of the event, by setting appropriate goals, clarifying conditions, and seeking corresponding solving algorithms; 3. With the help of computer artificial intelligence, the The situation of the event that occurs is used as the solution target, with the help of spatial information sharing service platform, GIS, data mining and other technologies, combined with emergency plan, emergency response knowledge base, case base, event information, and surrounding environment information, the emergency response plan is automatically generated. And adjust the plan through human-computer interaction, generate a scientific and reasonable disposal plan through the system, and test the rationality and effectiveness of the plan, so as to ensure that response measures can be taken quickly and effectively when an emergency occurs, and provide Provide support for the improvement of the plan.

When an emergency occurs, with the generation of a large amount of data, how can the data be analyzed in a short period of time, so that decision makers can intuitively understand the overall situation of the entire event, and quickly extract the relevant, feasible, and effective disposal results? Knowledge and improving the efficiency of dealing with emergencies are issues that must be considered by every emergency command system.

Contents of the invention

One of the problems to be solved by the present invention is to provide a method for constructing a knowledge base of an emergency auxiliary decision-making system based on location information, to realize the transformation of the historical knowledge base into the location knowledge base, the capture of webpage dynamic data and the dynamic growth of the location knowledge base, and to provide The knowledge information in the knowledge base increases the attribute of location information to associate knowledge with location information, and provides conditions for knowledge analysis and rapid retrieval based on location information.

One of the problems to be solved by the present invention is to provide a location information-based emergency auxiliary decision-making system knowledge base query method, which uses the knowledge base with location field information to retrieve information from the knowledge base according to the distance relationship between the emergency location and the knowledge location. The knowledge that is closest to the location where the emergency occurred makes the knowledge provided highly relevant, feasible, and effective in handling.

The technical scheme that the present invention solves one of above-mentioned technical problem is:

For the knowledge in the historical knowledge base or network knowledge that constitutes the knowledge source of the location information knowledge base, first extract the information and extract the address information; then analyze the address information to finally determine the heterogeneous effective address information; According to the information analysis results, the knowledge is classified according to the address coding specification, and the latitude and longitude information is assigned according to the address coding specification, and the relationship table between knowledge and location information is established in the database; thus a knowledge base based on location information is formed.

Real-time network knowledge is obtained from regular crawling of fixed web pages through web crawler tools according to keyword settings.

The extraction method is to split the content of the article by means of a binary tree, compare the split words with the words in the standard address library, if they are in the library, then extract them, otherwise filter them out.

When determining heterogeneous effective address information, the same name or alias analysis, address location analysis, and address hierarchy analysis are used; and it is defined as: the address information using the alias and the standard address information belong to the same address, and the address information is merged, The address in the title is more valuable than the address in the content, the address with high frequency is more valuable than the address with low frequency, and the high-level address is more valuable than the low-level address. The address level in the standard address library is divided into provinces layer, city layer, and township layer.

The association relationship includes the address of knowledge, the latitude and longitude information of knowledge, and the standard name of the address of knowledge; the priority order of knowledge classification is first position and then frequency, that is, if the position can be determined first, the address with dominant position will be used as the article If the final address of the article cannot be determined, the address with the highest frequency of occurrence will be used as the final address of the article.

The technical scheme that the present invention solves above-mentioned technical problem two is:

The main steps are:

Enter the retrieval conditions, that is, information keywords;

The client software on the mobile terminal sends information keywords and location information to the server;

The server first uses the keyword information to search, and uses the geographical information system to analyze the retrieved knowledge according to its location information, and returns it to the client;

The client software receives the above information, and displays the retrieved relevant knowledge information according to the location information.

Users select valid information to browse and make decisions on how to deal with emergencies.

There are two display methods, one is to mark the information on the map, and the other is to use the classification list.

First select the knowledge base search area, use geographic information tools to mark a rectangular, circular or polygonal area on the map, or check the address to select the area; then, follow the steps below;

Enter the retrieval conditions, that is, information keywords;

The client software on the mobile terminal sends information keywords and retrieval area information to the server;

The server first retrieves the knowledge in the region based on the regional information and using geographic information support software;

The server uses the keyword retrieval algorithm to perform keyword matching on the separated knowledge, and retrieves relevant knowledge information;

The server returns the search results to the client through the network, including information relevance priority, information location information, etc.;

The client software receives the above information, and displays the retrieved relevant knowledge information according to the location information;

Users select valid information to browse and make decisions on emergencies.

According to the general regional characteristics of emergencies, the present invention uses the association between knowledge and location information to develop a set of emergency auxiliary decision-making system based on location information retrieval. The system can classify and display information according to location correlation, providing Provide users with data closely related to the location of emergencies, which strongly supports the efficiency of emergency response to emergencies. The present invention comprehensively utilizes emergency event location information, geographic information system and knowledge base technology, so that the retrieved knowledge information is close to the emergency event location information, has high correlation, strong feasibility, good disposal effect, and strong support the work of decision makers.

Description of drawings

Below in conjunction with accompanying drawing, the present invention is further described:

Fig. 1 is an overall frame diagram of the present invention;

Fig. 2 is a flow chart of knowledge base construction of the present invention;

Fig. 3 is a flow chart of knowledge base retrieval in the present invention.

Detailed ways

As shown in FIG. 1 , the entire application architecture of the present invention is divided into four layers, which are respectively an application layer, an application support layer, a data layer and a basic support layer.

From the perspective of application, the emergency auxiliary decision-making system based on location information retrieval mainly includes two parts: the construction of auxiliary decision-making knowledge base and the application of emergency auxiliary decision-making. The knowledge base based on the location information is established through the auxiliary decision-making knowledge base construction software, and the knowledge in the knowledge base of the specific application location information is applied through the emergency auxiliary decision-making application.

Divided from the supporting function modules, the emergency auxiliary decision-making system based on location information retrieval mainly includes modules such as information extraction, information capture, knowledge classification, positioning information acquisition, knowledge retrieval, information analysis, geographic information system, and user management.

From the data layer, it mainly includes network real-time data, historical knowledge base, location knowledge base, basic geographic information base, business database, etc. When the network real-time data is specified, the knowledge related to the emergency management of emergencies is captured from the network, and the historical knowledge base stores the historical knowledge related to the emergency management of emergencies without location classification. The emergency auxiliary decision-making system based on location information retrieval has outstanding application effects on knowledge such as cases, laws and regulations, and emergency plans.

The basic support layer mainly includes mobile terminals, WEB servers, workstations, networks, middleware, GPS modules and GIS servers. Middleware is located on the heterogeneous hardware platform and operating system, it shields the differences of heterogeneous hardware platforms and operating systems, and provides services such as load balancing, transaction management, and reliable message delivery for upper-layer applications. The mobile terminal and workstation are the output terminals of the system information, and the GPS module is used to collect the user's current location information.

As shown in Figure 2, the establishment of the location knowledge base is the basis of the emergency decision-making system based on location information retrieval. As the name implies, the location knowledge base includes location, knowledge and the associated information between them. The knowledge sources of the location knowledge base include the knowledge in the historical knowledge base and the knowledge in the real-time data of the network. Through the information extraction, information analysis and information classification of the knowledge in these databases, the location knowledge base is finally formed. The conversion of historical knowledge base into location knowledge base belongs to static data, which is completed at one time under the operation of the administrator; the formation of location database from network real-time database belongs to the conversion of dynamic data, which requires the support of web crawlers and other tools to realize the conversion of location knowledge base. Dynamic growth of knowledge. The establishment of location knowledge base provides conditions for knowledge analysis and fast retrieval based on location information.

To sum up, the specific implementation scheme of knowledge base construction based on location information is as follows:

1) Knowledge base based on location information There are two sources of knowledge, one is historical knowledge base and the other is real-time network data. Real-time network data is the information regularly captured from fixed web pages through web crawler tool software and according to keyword settings.

2) For the knowledge in the historical knowledge base or the network knowledge, first extract the information and extract the address information. The extraction method is to split the content of the article by means of a binary tree, and compare the split words with those in the standard address library. If they are in the library, they are extracted, otherwise they are filtered out.

3) For the same article, multiple sets of address information may be extracted, and these address information need to be analyzed to finally determine heterogeneous effective address information. Analysis methods include address homonym or alias analysis, address location analysis, and address hierarchy analysis. The address information using the alias and the standard address information belong to the same address, and the address information needs to be merged; the address in the title is more valuable than the address in the content, the address with a large frequency of occurrence is more valuable than the address with a small frequency, and a high-level address It is more valuable than low-level addresses; the address levels in the standard address database are divided into province level, city level, and township level.

4) According to the information analysis results, classify the knowledge according to the address coding specification, and assign the latitude and longitude information according to the address coding specification, and establish a relationship table between knowledge and location information in the database. The relationship includes the address of knowledge, knowledge Longitude and latitude information, knowledge of the address standard name. The priority order of knowledge classification is position first and then frequency, that is, if the position priority can be determined, the address with dominant position will be used as the final address of the article; if the position priority cannot be determined, the address with high frequency of occurrence will be used as the address The final address of the article.

5) Form a knowledge base based on location information.

In the above process, the information classification adopted is to use the standard address database, which contains the address information and alias information of all provinces, cities and towns in China, and the latitude and longitude information of the administrative center of the address.

A lot of knowledge, including knowledge of historical cases, knowledge of laws and regulations, and knowledge of emergency plans, etc., has a great relationship with the region. Only when the region is close can it have better reference value, be closer to actual operation, and the validity and value of information will be higher. big. The knowledge retrieval and display method based on location information is a specific application of the location knowledge base. Through the mobile client software, mobile terminal users can query not only content-related data but also location-related data for analysis, which is conducive to the formation of effective decision-making. The spatial analysis function of the current geographic information system software is already very powerful, and it can quickly retrieve information in areas of arbitrary shapes including circles, rectangles, polygons, etc. from the database, providing a good foundation for the application of knowledge bases based on location information. Technical Support.

To sum up, as shown in Figure 3, the specific implementation scheme of the knowledge retrieval and display method based on location information is as follows:

1) Knowledge retrieval and display based on location information requires location knowledge base support and GIS software support. The knowledge base based on location information is the source of information, and GIS software is a key tool for rapid analysis and retrieval of information.

2) Hardware support and client software support are required. The hardware is mainly a mobile terminal and a GPS module. The mobile terminal plays the role of information input, information transmission, information reception, information display, etc., and uses the GPS module of the mobile terminal to obtain the current location. information.

3) Enter the retrieval conditions, that is, information keywords.

4) The client software on the mobile terminal sends information keywords and location information to the server.

5) The server first uses the keyword information to search, and uses the geographical information system to analyze the retrieved knowledge according to its location information, and returns it to the client.

6) The client software receives the above information, and displays the retrieved relevant knowledge information based on the location information. There are two ways to display, one is to mark the information on the map, and the other is to use the classification list. Marking on the map can intuitively display the characteristics of emergency information, and it is convenient to browse the information by using the classification list.

7) The user selects valid information to browse, and makes a decision on the handling of emergencies.

Knowledge retrieval based on location information also supports another way of application, that is, the user first selects and specifies an area, and then enters a search keyword. The information is displayed, and the specific implementation plan is:

1) To select the knowledge base search area, you can use geographic information software to mark a rectangular, circular or polygonal area on the map, or check the address to select the area.

2) Enter the retrieval conditions, that is, information keywords.

3) The client software on the mobile terminal sends information keywords and retrieval area information to the server.

4) The server first uses geographic information support software to retrieve knowledge in the region based on regional information.

5) The server uses the keyword retrieval algorithm to perform keyword matching on the separated knowledge to retrieve relevant knowledge information.

6) The server returns the search results to the client through the network, including information relevance priority, information location information, etc.

7) The client software receives the above information, and displays the retrieved relevant knowledge information based on the location information. There are two ways to display, one is to mark the information on the map, and the other is to use the classification list. Marking on the map can intuitively display the characteristics of emergency information, and it is convenient to browse the information by using the classification list.

8) Users select valid information to browse and make decisions on emergencies.

Claims (10)

1. The method for constructing the knowledge base of the emergency auxiliary decision-making system based on location information is characterized in that: for the knowledge in the historical knowledge base or the network knowledge that constitutes the knowledge source of the location information knowledge base, first carry out information extraction on it, and extract the Address information; then analyze the address information, and finally determine the heterogeneous effective address information; according to the information analysis results, classify the knowledge according to the address coding specification, and assign the latitude and longitude information according to the address coding specification, and establish knowledge in the database The association relationship table with location information; thus forming a knowledge base based on location information. 2. The knowledge base construction method according to claim 1, characterized in that: the real-time network knowledge is obtained by regularly grabbing from fixed web pages according to keyword settings through a web crawler tool. 3. The knowledge base construction method according to claim 1, characterized in that: the extraction method is to split the content of the article by means of a binary tree, and compare the split words with the words in the standard address base , if it is in the library, extract it, otherwise filter it out. 4. The knowledge base construction method according to claim 2, characterized in that: the extraction method is to split the content of the article by means of a binary tree, and compare the split words with the words in the standard address base , if it is in the library, extract it, otherwise filter it out. 5. The method for building a knowledge base according to any one of claims 1 to 4, characterized in that: when determining heterogeneous effective address information, analysis of address same name or alias, address location analysis, and address hierarchy analysis are used; It is defined as: the address information using an alias and the standard address information belong to the same address, and the address information is merged. The address in the title is more valuable than the address in the content, and the address with a large frequency of occurrence is more valuable than an address with a small frequency. High-level addresses are more valuable than low-level addresses. The address levels in the standard address database are divided into province level, city level, and township level. 6. The method for constructing a knowledge base according to any one of claims 1 to 4, characterized in that: the associated relationship includes the address of knowledge, the latitude and longitude information of knowledge, and the address standard name of knowledge; the priority order of knowledge classification is First position and then frequency, that is, if the position priority can be determined, the address with dominant position will be used as the final address of the article. If the position priority cannot be determined, the address with high frequency of occurrence will be used as the final address of the article. 7. The knowledge base construction method according to claim 5, characterized in that: said association relationship comprises the address of knowledge, the latitude and longitude information of knowledge, and the address standard name of knowledge; the priority order of knowledge classification is first position and then frequency, That is, if the location priority can be determined, the address with the dominant location will be used as the final address of the article. If the location priority cannot be determined, the address with the highest frequency of occurrence will be used as the final address of the article. 8. The knowledge base inquiry method that any one of claim 1-7 method construction is characterized in that: main step is: Enter the retrieval conditions, that is, information keywords; The client software on the mobile terminal sends information keywords and location information to the server; The server first uses the keyword information to search, and uses the geographical information system to analyze the retrieved knowledge according to its location information, and returns it to the client; The client software receives the above information, and displays the retrieved relevant knowledge information according to the location information. Users select valid information to browse and make decisions on how to deal with emergencies. 9. The knowledge base query method according to claim 8, characterized in that: there are two display methods, one is to mark information on the map, and the other is to use a classification list. 10. The knowledge base query method according to claim 8 or 9, characterized in that: first select the knowledge base retrieval area, use the geographic information tool to mark a rectangular, circular or polygonal area on the map, or tick the address Select, select the area; then, follow the steps below; Enter the retrieval conditions, that is, information keywords; The client software on the mobile terminal sends information keywords and retrieval area information to the server; The server first retrieves the knowledge in the region based on the regional information and using geographic information support software; The server uses the keyword retrieval algorithm to perform keyword matching on the separated knowledge, and retrieves relevant knowledge information; The server returns the search results to the client through the network, including information relevance priority, information location information, etc.; The client software receives the above information, and displays the retrieved relevant knowledge information according to the location information; Users select valid information to browse and make decisions on emergencies.

Images