CN109949088B

CN109949088B - Automatic mining analysis method for new shared economic project

Info

Publication number: CN109949088B
Application number: CN201910200109.2A
Authority: CN
Inventors: 张岩; 刘波; 董伟; 刘斌; 赵然; 姚鹏飞
Original assignee: Cisri Energy Saving Technology Co ltd
Current assignee: Cisri Energy Saving Technology Co ltd
Priority date: 2019-03-15
Filing date: 2019-03-15
Publication date: 2021-06-15
Anticipated expiration: 2039-03-15
Also published as: CN109949088A

Abstract

An automatic mining analysis system for a new shared economic project is characterized by comprising the following components: the system comprises a demand information server, a goods supply information server, a commodity paraphrasing server, a regional demand server group, a regional supply server group, an expert terminal group, a management terminal group and an authentication server. The automatic mining analysis method for the new shared economic project is carried out by using the system, and is characterized by comprising the following steps: 1) a preparation step; 2) information is captured and processed; 3) a sharing possibility mining step; 4) browsing and managing.

Description

Automatic mining analysis method for new shared economic project

Technical Field

The invention relates to the technical field of shared economic information analysis, in particular to an automatic mining and analyzing system and method for a new shared economic project.

Background

In the current information society, the new commodity demand is very fast, and the new supply information can also be fast found, if the commodity demand or the supply information with the approximately corresponding quantity exists in a similar area, the existence of the possibility of bridging the two leading lines and forming a new shared economic project is realized. The business method of projects such as automobile sharing and skill sharing is to link two service parties, and the shared economic project concerned by the application is mainly to grasp the demands of the two supply and demand parties of the existing commodity which are not realized or recognized by the two supply and demand parties, which actually generate a corresponding relation, to form a new shared economic project, so that the project has basic profitability feasibility.

At present, some analysis modes for sharing economic items exist in the prior art, such as capturing electric commerce search information for analysis, capturing shopping information, catering consumption information and the like for analysis, but the analysis in the prior art generally has two defects, one is mostly from result data analysis, the defect is very obvious, the result data is often far from the actual demand, for example, the result data means that the demand is met, or the result data is actually only a substitution mode which is not adopted because the consumer wants goods, the user demand, such as the data searched by the electric commerce, cannot be captured, and the important defect is that. Secondly, more importantly, without paying attention to the locality of the analysis data, if the data are placed nationwide, the shared economic project is different from the established conditions of the e-commerce sales, and the data are bound to be in certain areas, such as the Jingjin Ji area, the Changtriangle, Jiangzhe Shanghai and the like, such as the shared automobile, if the e-commerce is limited to driving in Jiangzhe Shanghai, the maintenance cost is different from the natural state of the allowed driving out limited area, which also causes whether the project is profitable, the existing e-commerce data analysis is often analyzed according to the correspondence between the manufactured products and the purchase requirements, and the analysis result is generally different from the requirements of the shared economic project as long as the e-commerce data are mailed, and cannot be used as effective reference data.

Yet another more prominent point is the timeliness of information gathering, for example, if a web page is run in a region such as region a and photographs are shared, selling and purchasing a full set of ingredients of the web page, perhaps a potential shared economic project, would have a large profit margin if the buyer and seller were connected in APP form, but this heat could quickly pass, and if the opportunity could not be found and the project taken at horse within a limited time, such as half a year, it would be difficult to take advantage of the popularity of the project to generate the maximum profit. In the prior art, some enterprises usually guess hot projects and make money, but the judgment is often carried out by means of commercial intuition instead of effective information reference, which means that the universality is not available.

Generally, the prior art lacks a corresponding analysis method and means for which shared items can be identified in the future. The prior art has no means for automatic analysis and provision, and the collection is carried out in the forms of workshop, questionnaire and the like, so that the cost is high, the timeliness is low, the bias is caused by a result theory, the logistics cost is too high due to no attention on the locality, the timeliness is insufficient, the effective opportunity is missed, and the pertinence is insufficient and is not comprehensive.

Disclosure of Invention

The first purpose of the invention is to solve the problem of the prior art that a regional and time-efficient mining method for sharing economic projects is lack, and the system can periodically monitor in real time and automatically generate suggestion reports when conditions are met, so that the system can provide auxiliary information very effectively. Of course, the scheme of the application only aims at the collection, calculation and supply of the information, how the information is perceived and utilized, how much the information is referred to, and the user determines the information and is not directly connected with the application.

An automatic mining analysis system for a new shared economic project is characterized by comprising the following components: the system server is provided with a system software platform which performs calculation, gives charts and verifies data output and input by the system server; the system server is provided with a first memory which is exclusively used for storing project feasibility reports; a defined category is selected in the system software platform.

The system comprises a demand information server, a plurality of electronic commerce information platforms and a plurality of electronic commerce information processing devices, wherein the demand information server is used for recording original commodity search information received from the electronic commerce information platforms, the original commodity search information is recorded in the demand information server one by one, and each piece of recorded original commodity search information at least comprises search text information, search position information and a search equipment name; and the original goods search information is limited to concern only information in a certain actual region; the contents which are not in the limited category are excluded through semantic analysis when the original commodity searching information is collected; the demand information server also has a second memory therein.

A goods supply information server for recording original goods supply information received from a goods supply website address confirmed by an expert terminal, wherein the original goods supply information is recorded in the goods supply information server one by one, and each piece of recorded original goods supply information at least comprises goods supply name information, goods supply quantity information and goods supply position information; and the original supply information is limited to concern only information in a certain actual region; the original goods supply information is collected, and the contents which are not in the limited category are eliminated through semantic analysis; the demand information server also has a third memory therein.

The commodity paraphrasing server is stored with a plurality of commodity paraphrasing cluster information under a limited category, and each piece of commodity paraphrasing cluster information comprises one formal name and a plurality of extension words of which the quantity can be continuously supplemented; the commodity paraphrase server is provided with semantic analysis engine software which can automatically extract nouns from text information and can generate commodity paraphrase cluster information from the extracted nouns according to semantic similarity.

The regional demand server group is provided with N regional demand servers, each regional demand server corresponds to different actual region ranges in a certain actual region, and the actual region ranges corresponding to any two regional demand servers are not overlapped or overlapped; n is a positive integer; the regional supply server group is provided with N regional supply servers, each regional supply server corresponds to different actual region ranges in a certain actual region, and the actual region ranges corresponding to any two regional supply servers are not overlapped or overlapped; n is a positive integer.

And the expert terminal group is provided with ZZ expert terminals and is used for collecting information from expert users and carrying out data interaction with the system server, the required information server, the goods supply information server, the commodity paraphrasing server, the regional required server group and the regional goods supply server group, and ZZ is a positive integer larger than or equal to 300.

The authentication server is provided with a prestored management personnel information group, the management personnel information group is provided with encrypted identification information which is stored in an entry mode according to the personnel ID number, and the encrypted identification information comprises but is not limited to reserved code information, fingerprint information, iris information and facial image information of personnel.

The management terminal group is composed of at least G management terminals, and each management terminal receives information in a wireless mode and performs data interaction with the system server, the demand information server, the goods supply information server, the commodity paraphrasing server, the regional demand server group, the regional goods supply server group and the authentication server; g is an integer greater than or equal to 100.

Preferably, the name of the searching device is an external name of the searching device on the internet, or a network card MAC address or a mobile phone identification code of the searching device; the demand information server and the supply information server are respectively provided with one or more demand mirror image servers and supply mirror image servers, and the demand mirror image server and the supply mirror image server continuously backup the newly added information of the demand information server and the supply information server at a time interval of 5 min; any one of the regional requirement servers and any one of the regional supply servers are arranged in the following two ways: all located in the same geographical location or each located within its corresponding physical area.

An automatic mining analysis method for a new shared economic project, which is performed by using the automatic mining analysis system for a new shared economic project, is characterized by comprising the following steps: 1) the preparation method comprises the following steps: determining the duration of a calculation period A and the duration of a calculation period B, wherein the duration of the calculation period B is integral multiple of the duration of the calculation period A; and utilizing semantic analysis engine software to extract nouns from the initial batch information of the original commodity search information and the initial batch information of the original commodity search information, and generating initial batch commodity paraphrase cluster information by all the extracted nouns according to the semantic similarity.

And generating a list of all initial batch commodity paraphrase cluster information, sending the list to all expert terminals, requiring each expert terminal to return suggested formal name information from each piece of the initial batch commodity paraphrase cluster information to a system server, sequencing all suggested formal name information returned by each piece of the initial batch commodity paraphrase cluster information according to the suggested number by the system server, and naming each piece of commodity paraphrase cluster information by taking the most suggested number as the formal name.

And confirming that the demand information server obtains information receiving authorization from the plurality of electronic commerce information platforms, testing the received commodity searching information aiming at each electronic commerce information platform, and sending the electronic commerce information platform information passing the test to a preset position of the demand information server for storage.

The system software platform sends prompts to all expert terminals, each expert terminal is required to provide at least one credible supply website address, the supply website address is a complete IP address or a domain name address, the system server sorts all supply website addresses returned by all the expert terminals according to the occurrence times, tests for capturing supply information are carried out on the supply website addresses with the occurrence times exceeding 50-80 times, and the supply website addresses passing the tests are used as initial batch supply website addresses and are sent to the supply information server for storage. Selecting a limited category, establishing a related phrase for the selected limited category by a system software platform, capturing commodity name information related to the limited category from the Internet, and filling the related phrase through similarity verification of the system software platform until the number of terms in the related phrase is not less than 2000; and transmitting the phrase to the affiliated demand information server and the supply information server for storage.

2) And information grabbing and processing step: setting a starting time, and starting from the starting time, performing the following actions in a first calculation period A: the demand information server receives original commodity search information generated in the calculation period A from an electronic commerce information platform passing the test, deletes original commodity search information with incomplete search text information, search position information and search equipment name information, deletes original commodity search information with search text information which is not similar to any word in a phrase, searches corresponding commodity definition information in the commodity definition server according to the search text information, records certain commodity definition cluster information corresponding to the commodity definition server as processing demand information XJ, and stores the processing demand information XJ in a storage position defined for the calculation period A; and sending any piece of commodity definition cluster information which cannot correspond to the commodity definition server to a storage position defined by the calculation period A in the second storage. The goods supply information server receives original goods supply information generated in the calculation period A from an initial batch goods supply website address, deletes the original goods supply information with incomplete goods supply name information, goods supply quantity information and goods supply position information, deletes the original goods supply information with goods supply name information being not similar to any word in the phrase, searches corresponding commodity definition information in the commodity definition server according to the goods supply name information, records certain commodity definition cluster information in the corresponding commodity definition server as processing goods supply information GJ and stores the processing goods supply information GJ at a storage position defined for the calculation period A; and sending any piece of commodity paraphrase cluster information which cannot correspond to the commodity paraphrase server to a storage position defined by the calculation period A in a third storage.

At the end of the first calculation cycle a, the following operations are carried out: utilizing semantic analysis engine software to extract nouns from all information in the original second memory and the original third memory, and generating commodity paraphrase cluster information of the calculation period A by all the extracted nouns according to semantic similarity; and generating a list of all the commodity paraphrase cluster information of the calculation period A, sending the list to all expert terminals, requiring each expert terminal to return information whether each item in the commodity paraphrase cluster information of the calculation period A is confirmed or not, giving suggested formal name information to the confirmed information, feeding the suggested formal name information back to the system server, recording the commodity paraphrase cluster information of the calculation period A with the confirmation number reaching or exceeding the QR value by the system server, naming each item of commodity paraphrase cluster information by the formal name with the maximum suggestion number, supplementing the commodity paraphrase cluster information into the plurality of commodity paraphrase cluster information, and emptying the second storage and the third storage at the moment.

Sending all processing demand information XJ recorded in the calculation period A in the demand information server to corresponding servers in N regional demand servers for storage according to different search position information; and sending all the processing goods supply information GJ recorded in the calculation period A in the goods supply information server to a corresponding server in the N regional goods supply servers for storage according to the difference of the search position information.

3) And (3) mining sharing possibility: continuously repeating the step (2) according to the time sequence to realize the calculation of the calculation period A one by one, and simultaneously dividing the N areas into M area groups, wherein M is a positive integer smaller than N; each regional group comprises one or more N regions, and the N regional demand servers and the N regional supply servers are also divided into M groups; at the end of each calculation cycle B, the following operations are carried out: calculating the regional demand server and regional supply server groups of all the groups in the M groups, wherein for the ith group in the 1 st-M groups, i is 1, … …, M; respectively summarizing all the processing demand information XJ and all the processing supply information GJ of the ith group in the calculation period B, if the matching condition simultaneously meeting the following A, B conditions occurs: A) the goods supply name information and the search text information belong to the same commodity paraphrase cluster information; B) for the machining demand information XJ and all machining supply information GJ meeting the condition A, 1-C is larger than sigma supply quantity information/sigma XJ is smaller than 1+ C; c is a constant between 0.05 and 0.3; recording the matching condition information of the ith group information, the commodity paraphrase cluster information, the actually matched processing demand information XJ and all processing supply information GJ when actual matching occurs; and generating shared project mining entries, after the regional demand servers and regional supply servers of all the groups in all the M groups are calculated, generating a current mining report of the calculation period B with a catalog, an abstract and a full text from all the shared project mining entries, and sending the current mining report to a reserved position of a system server for storage.

4) Browsing and managing: continuously repeating the steps (2) and (3) according to the time sequence, and continuously forming a new primary mining report at the end of each calculation period B; when a new first-stage mining report is formed, the summary of the first-stage mining report is sent to all management terminals, the management terminals which request the full-text reading request can be accepted, and the management terminals which receive the reading request simultaneously send display information and voice prompt information to request identity authentication.

And after the authentication is successful, the system software platform confirms and sends full text information of the mining report to the management terminal.

Preferably, the calculation period a is 1 week, and the calculation period B is 5 weeks; the value of QR is any positive integer from 20 to 50; c is a constant between 0.1 and 0.2.

The invention has the advantages that the invention can be mainly divided into five points, firstly, aiming at the time-critical information society, a whole set of system can replace on-site workshop, joint questionnaire survey, special evaluation activities or mass calling to obtain effective analysis information, greatly save manpower and material resources, continuously obtain information in each period, capture whether business opportunities related to shared economy exist in real time, and realize the most intuitive reference information of a commodity research and development plan. Secondly, from the system structure, the regional analysis is realized, because if the project does not belong to the same region, the project can not be implemented due to the logistics time and the money cost, and the matching and the guarantee can not be carried out. The prior art does not have such analytical means. The M groups can be selected at will, so that attention to all areas with low logistics cost and convenient traffic is realized, and the method is an effective analysis means. And thirdly, the administrator group can conveniently obtain the calculated information, the query is convenient, and reasonable query information can be obtained by different management levels according to the preset authority, so that the information is conveniently and fully utilized in time.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.

FIG. 1 is a flow chart of a new shared economic project automatic mining analysis method of the present invention.

Detailed Description

The preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings so that the advantages and features of the present invention can be more easily understood by those skilled in the art, and the scope of the present invention will be more clearly and clearly defined.

Example 1

In this embodiment, the region analyzed here is, for example, nationwide, and the limited categories we are concerned with here are, for example, the fields of vehicles excluding automobiles, such as electric mopeds, shared bicycles, scooters, and various non-automotive vehicles, and so on.

The system comprises a demand information server, a plurality of electronic commerce information platforms and a plurality of electronic commerce information processing devices, wherein the demand information server is used for recording original commodity search information received from the electronic commerce information platforms, the original commodity search information is recorded in the demand information server one by one, and each piece of recorded original commodity search information at least comprises search text information, search position information and a search equipment name; and the original goods search information is limited to concern only information in a certain actual region; the contents which are not in the limited category are excluded through semantic analysis when the original commodity searching information is collected; the demand information server also has a second memory therein. In order to ensure the efficiency of analysis, the search information and other information to be analyzed may be set in a text format. The semantic analysis here can exclude information that is not likely to be a vehicle, as follows.

The commodity paraphrasing server is stored with a plurality of commodity paraphrasing cluster information under a limited category, and each piece of commodity paraphrasing cluster information comprises one formal name and a plurality of extension words of which the quantity can be continuously supplemented; the commodity paraphrase server is provided with semantic analysis engine software which can automatically extract nouns from text information and can generate commodity paraphrase cluster information from the extracted nouns according to semantic similarity. Here, the term "commodity paraphrase cluster information" is, for example, an electric scooter, and the terms of extension include, for example, an electric skateboard, an electric scooter, a skateboard with electricity, an electrified skateboard, and the like.

The regional demand server group is provided with N regional demand servers, each regional demand server corresponds to different actual region ranges in a certain actual region, and the actual region ranges corresponding to any two regional demand servers are not overlapped or overlapped; n is a positive integer; the regional supply server group is provided with N regional supply servers, each regional supply server corresponds to different actual region ranges in a certain actual region, and the actual region ranges corresponding to any two regional supply servers are not overlapped or overlapped; n is a positive integer. Here, the analysis of the demand and supply is the same for each of the N regions, so that the analysis has correspondence, for example, all correspond to several neighboring cities, a certain municipality or province.

The management terminal group is composed of at least G management terminals, and each management terminal receives information in a wireless mode and performs data interaction with the system server, the demand information server, the goods supply information server, the commodity paraphrasing server, the regional demand server group and the regional goods supply server group authentication server; g is an integer greater than or equal to 100.

Preferably, the name of the searching device is an external name of the searching device on the internet, or a network card MAC address or a mobile phone identification code of the searching device; the demand information server and the supply information server are respectively provided with one or more demand mirror image servers and supply mirror image servers, and the demand mirror image server and the supply mirror image server continuously backup the newly added information of the demand information server and the supply information server at a time interval of 5 min; any one of the regional requirement servers and any one of the regional supply servers are arranged in the following two ways: all located in the same geographical location or each located within its corresponding physical area. Here, if information is collected from a computer or a pad, a palm, or the like, a MAC address may be registered, if the information is collected from a mobile phone, an IMEI code may be registered, and due to the repeatability of the same user, repeated information from the same user within a certain time interval may be registered as one piece of repeated information, for example, the same user searches for a plurality of times within one hour, but all records as one piece of information.

An automatic mining analysis method for a new shared economic project, which is performed by using the automatic mining analysis system for a new shared economic project, is characterized by comprising the following steps: 1) the preparation method comprises the following steps: determining the duration of a calculation period A and the duration of a calculation period B, wherein the duration of the calculation period B is integral multiple of the duration of the calculation period A; and utilizing semantic analysis engine software to extract nouns from the initial batch information of the original commodity search information and the initial batch information of the original commodity search information, and generating initial batch commodity paraphrase cluster information by all the extracted nouns according to the semantic similarity. The calculation period A is designed to be shorter to relieve the pressure of a supply and demand server, and the calculation period B is several times of the calculation period A, so that the calculation is valuable once, and the corresponding information in a short time is possibly not valuable and can not reflect the characteristics in a period of time.

And generating a list of all initial batch commodity paraphrase cluster information, sending the list to all expert terminals, requiring each expert terminal to return suggested formal name information from each piece of the initial batch commodity paraphrase cluster information to a system server, sequencing all suggested formal name information returned by each piece of the initial batch commodity paraphrase cluster information according to the suggested number by the system server, and naming each piece of commodity paraphrase cluster information by taking the most suggested number as the formal name. The rule for selecting the formal name is to make the paraphrase cluster information be correctly recognized by most people and facilitate information management.

The system software platform sends prompts to all expert terminals, each expert terminal is required to provide at least one credible supply website address, the supply website address is a complete IP address or a domain name address, the system server sorts all supply website addresses returned by all the expert terminals according to the occurrence times, tests for capturing supply information are carried out on the supply website addresses with the occurrence times exceeding 60 times, and the supply website addresses passing the tests are used as initial batch supply website addresses and are sent to a supply information server for storage. Selecting a limited category, establishing a related phrase for the selected limited category by a system software platform, capturing commodity name information related to the limited category from the Internet, and filling the related phrase through similarity verification of the system software platform until the number of terms in the related phrase is not less than 2500; and transmitting the phrase to the affiliated demand information server and the supply information server for storage.

2) And information grabbing and processing step: setting a starting time, and starting from the starting time, performing the following actions in a first calculation period A: the demand information server receives original commodity search information generated in the calculation period A from an electronic commerce information platform passing the test, deletes original commodity search information with incomplete search text information, search position information and search equipment name information, deletes original commodity search information with search text information which is not similar to any word in a phrase, searches corresponding commodity definition information in the commodity definition server according to the search text information, records certain commodity definition cluster information corresponding to the commodity definition server as processing demand information XJ, and stores the processing demand information XJ in a storage position defined for the calculation period A; and sending any piece of commodity definition cluster information which cannot correspond to the commodity definition server to a storage position defined by the calculation period A in the second storage. The goods supply information server receives original goods supply information generated in the calculation period A from an initial batch goods supply website address, deletes the original goods supply information with incomplete goods supply name information, goods supply quantity information and goods supply position information, deletes the original goods supply information with goods supply name information being not similar to any word in the phrase, searches corresponding commodity definition information in the commodity definition server according to the goods supply name information, records certain commodity definition cluster information in the corresponding commodity definition server as processing goods supply information GJ and stores the processing goods supply information GJ at a storage position defined for the calculation period A; and sending any piece of commodity paraphrase cluster information which cannot correspond to the commodity paraphrase server to a storage position defined by the calculation period A in a third storage. Here, XJ actually corresponds to a demand, but GJ actually includes its supply capability, for example, a certain foundries that 80 electric scooters can be produced.

At the end of the first calculation cycle a, the following operations are carried out: utilizing semantic analysis engine software to extract nouns from all information in the original second memory and the original third memory, and generating commodity paraphrase cluster information of the calculation period A by all the extracted nouns according to semantic similarity; and generating a list of all the commodity paraphrase cluster information of the calculation period A, sending the list to all expert terminals, requiring each expert terminal to return information whether each item in the commodity paraphrase cluster information of the calculation period A is confirmed or not, giving suggested formal name information to the confirmed information, feeding the suggested formal name information back to the system server, recording the commodity paraphrase cluster information of the calculation period A with the confirmation number reaching or exceeding the QR value by the system server, naming each item of commodity paraphrase cluster information by the formal name with the maximum suggestion number, supplementing the commodity paraphrase cluster information into the plurality of commodity paraphrase cluster information, and emptying the second storage and the third storage at the moment. Here, the confirmation number is, for example, greater than QR 30, and a certain threshold is set in order to avoid search information that does not show the intention of the crowd.

3) And (3) mining sharing possibility: continuously repeating the step (2) according to the time sequence to realize the calculation of the calculation period A one by one, and simultaneously dividing the N areas into M area groups, wherein M is a positive integer smaller than N; each regional group comprises one or more N regions, and the N regional demand servers and the N regional supply servers are also divided into M groups; at the end of each calculation cycle B, the following operations are carried out: calculating the regional demand server and regional supply server groups of all the groups in the M groups, wherein for the ith group in the 1 st-M groups, i is 1, … …, M; respectively summarizing all the processing demand information XJ and all the processing supply information GJ of the ith group in the calculation period B, if the matching condition simultaneously meeting the following A, B conditions occurs: A) the goods supply name information and the search text information belong to the same commodity paraphrase cluster information; B) for the machining demand information XJ and all machining supply information GJ meeting the condition A, 1-C is larger than sigma supply quantity information/sigma XJ is smaller than 1+ C; c is a constant of 0.3; recording the matching condition information of the ith group information, the commodity paraphrase cluster information, the actually matched processing demand information XJ and all processing supply information GJ when actual matching occurs; and generating shared project mining entries, after the regional demand servers and regional supply servers of all the groups in all the M groups are calculated, generating a current mining report of the calculation period B with a catalog, an abstract and a full text from all the shared project mining entries, and sending the current mining report to a reserved position of a system server for storage. The condition a means that the items are corresponding, for example, all the items need electric scooters, and the condition B means that the demand amount corresponds, the simplest example is, for example, the calculation result is that the demand of 8000 electric scooters appears in the first five weeks of a certain city, and the supply capacity of 7500 electric scooters appears at the same time, that is, the condition B is also satisfied, and then the project of proposing a shared skateboard in the city can be considered, which is a feasible plan.

Preferably, the calculation period a is 1 week, and the calculation period B is 5 weeks.

Example 2

In this implementation, the territory analyzed here is, for example, nationwide, and the defined categories we are interested in here are, for example, durable everyday items such as umbrellas, crutches, sun caps, hiking sticks, sunglasses, backpacks, and so on.

The system comprises a demand information server, a plurality of electronic commerce information platforms and a plurality of electronic commerce information processing devices, wherein the demand information server is used for recording original commodity search information received from the electronic commerce information platforms, the original commodity search information is recorded in the demand information server one by one, and each piece of recorded original commodity search information at least comprises search text information, search position information and a search equipment name; and the original goods search information is limited to concern only information in a certain actual region; the contents which are not in the limited category are excluded through semantic analysis when the original commodity searching information is collected; the demand information server also has a second memory therein. In order to ensure the efficiency of analysis, the search information and other information to be analyzed may be set in a text format. The semantic analysis here can exclude information that is not likely to be a durable commodity, as follows.

The commodity paraphrasing server is stored with a plurality of commodity paraphrasing cluster information under a limited category, and each piece of commodity paraphrasing cluster information comprises one formal name and a plurality of extension words of which the quantity can be continuously supplemented; the commodity paraphrase server is provided with semantic analysis engine software which can automatically extract nouns from text information and can generate commodity paraphrase cluster information from the extracted nouns according to semantic similarity. Here, the explanation group information of a commercial product is, for example, an umbrella, and the formal name includes, for example, a parasol, umbrella, and the like.

And the expert terminal group is provided with ZZ expert terminals and is used for collecting information from expert users and carrying out data interaction with the system server, the required information server, the goods supply information server, the commodity paraphrasing server, the regional required server group and the regional goods supply server group, and ZZ is a positive integer larger than or equal to 350.

The management terminal group is composed of at least G management terminals, and each management terminal receives information in a wireless mode and performs data interaction with the system server, the demand information server, the goods supply information server, the commodity paraphrasing server, the regional demand server group and the regional goods supply server group authentication server; g is an integer greater than or equal to 120.

Preferably, the name of the searching device is an external name of the searching device on the internet, or a network card MAC address or a mobile phone identification code of the searching device; the demand information server and the supply information server are respectively provided with one or more demand mirror image servers and supply mirror image servers, and the demand mirror image server and the supply mirror image server continuously backup the newly added information of the demand information server and the supply information server at a time interval of less than 3 min; any one of the regional requirement servers and any one of the regional supply servers are arranged in the following two ways: all located in the same geographical location or each located within its corresponding physical area. Here, if information is collected from a computer or a pad, a palm, or the like, a MAC address may be registered, if the information is collected from a mobile phone, an IMEI code may be registered, and due to the repeatability of the same user, repeated information from the same user within a certain time interval may be registered as one piece of repeated information, for example, the same user searches for a plurality of times within two hours, but all records as one piece of information.

The system software platform sends prompts to all expert terminals, each expert terminal is required to provide at least one credible supply website address, the supply website address is a complete IP address or a domain name address, the system server sorts all supply website addresses returned by all the expert terminals according to the occurrence times, tests for capturing supply information are carried out on the supply website addresses with the occurrence times exceeding 70 times, and the supply website addresses passing the tests are used as initial batch supply website addresses and are sent to a supply information server for storage. Selecting a limited category, establishing a phrase for the selected limited category by a system software platform, capturing commodity name information related to the limited category from the Internet, and filling the phrase through similarity verification of the system software platform until the number of terms in the phrase is not less than 2800; and transmitting the phrase to the affiliated demand information server and the supply information server for storage.

2) And information grabbing and processing step: setting a starting time, and starting from the starting time, performing the following actions in a first calculation period A: the demand information server receives original commodity search information generated in the calculation period A from an electronic commerce information platform passing the test, deletes original commodity search information with incomplete search text information, search position information and search equipment name information, deletes original commodity search information with search text information which is not similar to any word in a phrase, searches corresponding commodity definition information in the commodity definition server according to the search text information, records certain commodity definition cluster information corresponding to the commodity definition server as processing demand information XJ, and stores the processing demand information XJ in a storage position defined for the calculation period A; and sending any piece of commodity definition cluster information which cannot correspond to the commodity definition server to a storage position defined by the calculation period A in the second storage. The goods supply information server receives original goods supply information generated in the calculation period A from an initial batch goods supply website address, deletes the original goods supply information with incomplete goods supply name information, goods supply quantity information and goods supply position information, deletes the original goods supply information with goods supply name information being not similar to any word in the phrase, searches corresponding commodity definition information in the commodity definition server according to the goods supply name information, records certain commodity definition cluster information in the corresponding commodity definition server as processing goods supply information GJ and stores the processing goods supply information GJ at a storage position defined for the calculation period A; and sending any piece of commodity paraphrase cluster information which cannot correspond to the commodity paraphrase server to a storage position defined by the calculation period A in a third storage. Here, XJ actually corresponds to a demand, but GJ actually contains its supply capacity, e.g. a factory claims that it can produce 120 umbrellas.

At the end of the first calculation cycle a, the following operations are carried out: utilizing semantic analysis engine software to extract nouns from all information in the original second memory and the original third memory, and generating commodity paraphrase cluster information of the calculation period A by all the extracted nouns according to semantic similarity; and generating a list of all the commodity paraphrase cluster information of the calculation period A, sending the list to all expert terminals, requiring each expert terminal to return information whether each item in the commodity paraphrase cluster information of the calculation period A is confirmed or not, giving suggested formal name information to the confirmed information, feeding the suggested formal name information back to the system server, recording the commodity paraphrase cluster information of the calculation period A with the confirmation number reaching or exceeding the QR value by the system server, naming each item of commodity paraphrase cluster information by the formal name with the maximum suggestion number, supplementing the commodity paraphrase cluster information into the plurality of commodity paraphrase cluster information, and emptying the second storage and the third storage at the moment. Here, the confirmation number is, for example, greater than QR 40, and a certain threshold is set in order to avoid search information that does not show the intention of the crowd.

3) And (3) mining sharing possibility: continuously repeating the step (2) according to the time sequence to realize the calculation of the calculation period A one by one, and simultaneously dividing the N areas into M area groups, wherein M is a positive integer smaller than N; each regional group comprises one or more N regions, and the N regional demand servers and the N regional supply servers are also divided into M groups; at the end of each calculation cycle B, the following operations are carried out: calculating the regional demand server and regional supply server groups of all the groups in the M groups, wherein for the ith group in the 1 st-M groups, i is 1, … …, M; respectively summarizing all the processing demand information XJ and all the processing supply information GJ of the ith group in the calculation period B, if the matching condition simultaneously meeting the following A, B conditions occurs: A) the goods supply name information and the search text information belong to the same commodity paraphrase cluster information; B) for the machining demand information XJ and all machining supply information GJ meeting the condition A, 1-C is larger than sigma supply quantity information/sigma XJ is smaller than 1+ C; c is a constant of 0.25; recording the matching condition information of the ith group information, the commodity paraphrase cluster information, the actually matched processing demand information XJ and all processing supply information GJ when actual matching occurs; and generating shared project mining entries, after the regional demand servers and regional supply servers of all the groups in all the M groups are calculated, generating a current mining report of the calculation period B with a catalog, an abstract and a full text from all the shared project mining entries, and sending the current mining report to a reserved position of a system server for storage. The condition A means that the articles correspond to each other, for example, the umbrellas are needed, the condition B means that the quantity of the needed articles corresponds to each other, and the simplest example is that the calculation result is that the requirement of 6000 sun umbrellas appears in the first ten weeks of a certain city, and the supply capacity of 5800 sun umbrellas appears, namely the condition B is also met, and then a project of sharing the skateboard in the city can be considered and is a feasible plan.

Preferably, the calculation period a is 2 weeks, and the calculation period B is 10 weeks.

The above description is only an embodiment of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions that are not thought of through the inventive work should be included in the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope defined by the claims.

Claims

1. An automatic mining analysis method for a new shared economic project, which is performed by using an automatic mining analysis system for a new shared economic project, the automatic mining analysis system for a new shared economic project comprising:

the system server is provided with a system software platform which performs calculation, gives charts and verifies data output and input by the system server; the system server is provided with a first memory which is exclusively used for storing project feasibility reports; selecting a defined category in a system software platform;

the system comprises a demand information server, a plurality of electronic commerce information platforms and a plurality of electronic commerce information processing devices, wherein the demand information server is used for recording original commodity search information received from the electronic commerce information platforms, the original commodity search information is recorded in the demand information server one by one, and each piece of recorded original commodity search information at least comprises search text information, search position information and a search equipment name; and the original goods search information is limited to concern only information in a certain actual region; the contents which are not in the limited category are excluded through semantic analysis when the original commodity searching information is collected; the demand information server is also provided with a second memory;

a goods supply information server for recording original goods supply information received from a goods supply website address confirmed by an expert terminal, wherein the original goods supply information is recorded in the goods supply information server one by one, and each piece of recorded original goods supply information at least comprises goods supply name information, goods supply quantity information and goods supply position information; and the original supply information is limited to concern only information in a certain actual region; the original goods supply information is collected, and the contents which are not in the limited category are eliminated through semantic analysis; the demand information server is also provided with a third memory;

the commodity paraphrasing server is stored with a plurality of commodity paraphrasing cluster information under a limited category, and each piece of commodity paraphrasing cluster information comprises one formal name and a plurality of extension words of which the quantity can be continuously supplemented; the commodity paraphrase server is provided with semantic analysis engine software which can automatically extract nouns from text information and can generate commodity paraphrase cluster information from the extracted nouns according to semantic similarity;

the regional demand server group is provided with N regional demand servers, each regional demand server corresponds to different actual region ranges in a certain actual region, and the actual region ranges corresponding to any two regional demand servers are not overlapped or overlapped; n is a positive integer;

the regional supply server group is provided with N regional supply servers, each regional supply server corresponds to different actual region ranges in a certain actual region, and the actual region ranges corresponding to any two regional supply servers are not overlapped or overlapped; n is a positive integer;

the expert terminal group is provided with ZZ expert terminals and is used for collecting information from expert users and carrying out data interaction with the system server, the required information server, the goods supply information server, the commodity paraphrasing server, the regional required server group and the regional goods supply server group, and ZZ is a positive integer which is more than or equal to 300;

the authentication server is provided with a prestored management personnel information group, the management personnel information group is provided with encrypted identification information which is stored in an item manner according to a personnel ID number, and the encrypted identification information comprises but is not limited to reserved code information, fingerprint information, iris information and facial image information of personnel;

the management terminal group is composed of at least G management terminals, and each management terminal receives information in a wireless mode and performs data interaction with the system server, the demand information server, the goods supply information server, the commodity paraphrasing server, the regional demand server group, the regional goods supply server group and the authentication server; g is an integer greater than or equal to 100;

the searching equipment name is specifically an external name of the equipment for searching on the Internet, or a network card MAC address or a mobile phone identification code of the equipment for searching;

the demand information server and the supply information server are respectively provided with one or more demand mirror image servers and supply mirror image servers, and the demand mirror image server and the supply mirror image server continuously backup the newly added information of the demand information server and the supply information server at a time interval of 5 min;

any one of the regional requirement servers and any one of the regional supply servers are arranged in the following two ways: all the devices are arranged at the same geographical position or are respectively arranged in the corresponding actual area range;

the automatic mining analysis method for the new shared economic project comprises the following steps:

1) the preparation method comprises the following steps: determining the duration of a calculation period A and the duration of a calculation period B, wherein the duration of the calculation period B is integral multiple of the duration of the calculation period A; utilizing semantic analysis engine software to extract nouns from the initial batch information of the original commodity search information and the initial batch information of the original commodity search information, and generating initial batch commodity paraphrase cluster information by all the extracted nouns according to semantic similarity;

generating a list of all initial batch of commodity paraphrase cluster information, sending the list to all expert terminals, requiring each expert terminal to return suggested formal name information from each piece of the initial batch of commodity paraphrase cluster information to a system server, sequencing all suggested formal name information returned by each piece of the initial batch of commodity paraphrase cluster information according to the suggested number by the system server, and naming each piece of commodity paraphrase cluster information by taking the most suggested number as the formal name;

confirming that the demand information server obtains information receiving authorization from the plurality of electronic commerce information platforms, testing the received commodity searching information aiming at each electronic commerce information platform, and sending the electronic commerce information platform information passing the test to a preset position of the demand information server for storage;

the system software platform sends prompts to all expert terminals, each expert terminal is required to provide at least one credible supply website address, the supply website address is a complete IP address or a domain name address, the system server sorts all supply website addresses returned by all the expert terminals according to the occurrence times, tests for capturing supply information are carried out on the supply website addresses with the occurrence times exceeding 50-80 times, and the supply website addresses passing the tests are used as initial batch supply website addresses and are sent to a supply information server for storage;

selecting a limited category, establishing a related phrase for the selected limited category by a system software platform, capturing commodity name information related to the limited category from the Internet, and filling the related phrase through similarity verification of the system software platform until the number of terms in the related phrase is not less than 2000; transmitting the phrase to the request information server and the supply information server for storage;

2) and information grabbing and processing step: setting a starting time, and starting from the starting time, performing the following actions in a first calculation period A: the demand information server receives original commodity search information generated in the calculation period A from an electronic commerce information platform passing the test, deletes original commodity search information with incomplete search text information, search position information and search equipment name information, deletes original commodity search information with search text information which is not similar to any word in a phrase, searches corresponding commodity definition information in the commodity definition server according to the search text information, records certain commodity definition cluster information corresponding to the commodity definition server as processing demand information XJ, and stores the processing demand information XJ in a storage position defined for the calculation period A; sending any piece of commodity paraphrasing cluster information which cannot correspond to the commodity paraphrasing server to a storage position defined for the calculation cycle A in a second storage;

the goods supply information server receives original goods supply information generated in the calculation period A from an initial batch goods supply website address, deletes the original goods supply information with incomplete goods supply name information, goods supply quantity information and goods supply position information, deletes the original goods supply information with goods supply name information being not similar to any word in the phrase, searches corresponding commodity definition information in the commodity definition server according to the goods supply name information, records certain commodity definition cluster information in the corresponding commodity definition server as processing goods supply information GJ and stores the processing goods supply information GJ at a storage position defined for the calculation period A; sending any piece of commodity paraphrasing cluster information which cannot correspond to the commodity paraphrasing server to a storage position defined for the calculation cycle A in a third storage;

at the end of the first calculation cycle a, the following operations are carried out:

utilizing semantic analysis engine software to extract nouns from all information in the original second memory and the original third memory, and generating commodity paraphrase cluster information of the calculation period A by all the extracted nouns according to semantic similarity; generating a list of all the commodity paraphrase cluster information of the calculation period A, sending the list to all expert terminals, requiring each expert terminal to return information whether to confirm each item of the commodity paraphrase cluster information of the calculation period A, giving suggested formal name information to the confirmed information, feeding the suggested formal name information back to a system server, recording the commodity paraphrase cluster information of the calculation period A, the confirmation number of which reaches or exceeds a QR value, naming each item of commodity paraphrase cluster information by the formal name with the maximum suggestion number, supplementing the commodity paraphrase cluster information into the plurality of commodity paraphrase cluster information, and emptying the second storage and the third storage;

sending all processing demand information XJ recorded in the calculation period A in the demand information server to corresponding servers in N regional demand servers for storage according to different search position information; sending all the processing goods supply information GJ recorded in the calculation period A in the goods supply information server to a corresponding server in N regional goods supply servers for storage according to the difference of the search position information of the processing goods supply information GJ;

3) and (3) mining sharing possibility: continuously repeating the step (2) according to the time sequence to realize the calculation of the calculation period A one by one, and simultaneously dividing the N areas into M area groups, wherein M is a positive integer smaller than N; each regional group comprises one or more N regions, and the N regional demand servers and the N regional supply servers are also divided into M groups;

at the end of each calculation cycle B, the following operations are carried out:

calculating the regional demand server and regional supply server groups of all the groups in the M groups, wherein for the ith group in the 1 st-M groups, i is 1, … …, M;

respectively summarizing all the processing demand information XJ and all the processing supply information GJ of the ith group in the calculation period B, if the matching condition simultaneously meeting the following A, B conditions occurs:

A) the goods supply name information and the search text information belong to the same commodity paraphrase cluster information;

B) for the machining demand information XJ and all machining supply information GJ meeting the condition A, 1-C is larger than sigma supply quantity information/sigma XJ is smaller than 1+ C; c is a constant between 0.05 and 0.3;

recording the matching condition information of the ith group information, the commodity paraphrase cluster information, the actually matched processing demand information XJ and all processing supply information GJ when actual matching occurs; generating sharing project mining entries, after computing all the grouped regional demand servers and regional supply servers in all the M groups, generating current mining reports of the computing cycle B with catalogs, abstracts and full texts from all the sharing project mining entries, and sending the current mining reports to a reserved position of a system server for storage;

4) browsing and managing: continuously repeating the steps (2) and (3) according to the time sequence, and continuously forming a new primary mining report at the end of each calculation period B; when a new first-stage mining report is formed, the summary of the first-stage mining report is sent to all management terminals, the management terminals which request the full-text reading request can be accepted, and the management terminals which receive the reading request simultaneously send display information and voice prompt information to request identity authentication; and after the authentication is successful, the system software platform confirms and sends full text information of the mining report to the management terminal.

2. A method for automatically mining and analyzing a new shared economic project as claimed in claim 1, wherein:

the calculation period A is 1 week, and the calculation period B is 5 weeks;

the value of QR is any positive integer from 20 to 50;

c is a constant between 0.1 and 0.2.