CN112131459A - Intellectual property information retrieval software management system and method based on big data - Google Patents

Abstract

The invention discloses an intellectual property information retrieval software management system and method based on big data, comprising a data acquisition module, several user terminals, a browsing module, a user behavior analysis module, a server, a retrieval module, a data acquisition module, a database and an evaluation module; The retrieval module is used for the user terminal to issue the retrieval information and send the retrieval information to the server; the data acquisition module is used to collect the basic information of each patent and transmit the basic information of each patent to the database for storage; the investigation module is used to monitor every The communication information between the inventor of the patent and the user at the preset time and the communication information is transmitted to the data analysis module; the data analysis module receives the click information and communication information, and combines the retrieval module to carry out the push analysis of the patent. The invention cleverly utilizes big data intelligent analysis and user's behavior to improve search efficiency and reduce user's burden.

Description

Intellectual property information retrieval software management system and method based on big data

technical field

The invention relates to the field of information retrieval, in particular to a software management system and method for intellectual property information retrieval based on big data.

Background technique

With the popularization of Internet applications and the arrival of the era of big data, the number of Internet pages around the world increases by tens of millions every day. To retrieve the required information on the vast web, search engines have become an indispensable assistant for accessing the Internet.

The document of publication number CN106503199A discloses a network-based computer information retrieval system, including a foreground information input system and a background information retrieval system, and the foreground information input system and the background information retrieval system are both bidirectionally electrically connected through a computer center system; The foreground information input system includes a picture input subsystem, a language input subsystem and a text input subsystem; the background information retrieval system includes an information retrieval subsystem, a retrieval retrieval subsystem and a retrieval sharing subsystem. The computer information retrieval system of the network, including the foreground information input system and the background information retrieval system, can input three kinds of retrieval information: picture, language and text when retrieval is required, which overcomes the problem of single retrieval method of the traditional retrieval system, and the retrieval and sharing The subsystem realizes the sharing of retrieval and remote transmission.

However, this patent presents all possible results to the user, and the user selects the required search items; it increases the user's burden and reduces the search efficiency; and does not fully consider the user's behavior when sorting the search items.

SUMMARY OF THE INVENTION

In view of the deficiencies in the prior art, the purpose of the present invention is to provide a software management system and method for intellectual property information retrieval based on big data; the present invention cleverly utilizes big data intelligent analysis and user behavior to improve search efficiency and reduce user burden; The retrieval service system forms an effective evaluation, which is convenient for later viewing.

The object of the present invention can be realized through the following technical solutions:

An intellectual property information retrieval software management system based on big data, including a data acquisition module, several user terminals, a browsing module, a user behavior analysis module, a server, a retrieval module, a data acquisition module, a database and an evaluation module;

The user terminal is used to input the user's login information and registration information. The user logs in after entering the login information through the user terminal when there is an existing account. When the user does not have an account, the user enters the registration information through the user terminal and registers a new account. Log in;

The retrieval module is used for the user terminal to issue retrieval information and send the retrieval information to the server, where the retrieval information includes keywords and technical fields;

The data acquisition module is used to collect the basic information of each patent and transmit the basic information of each patent to the database for storage; the database is used to store the browsing records, evaluation records, retrieval information, login information and registration received by the server information;

The access statistics module is used to count the click information of each patent in the database within 10 days before the current time of the system and transmit the click information to the data analysis module; the investigation module is used to monitor the inventors and inventors of the patent every preset time. Communication information of users and transmission of communication information to the data analysis module;

The data analysis module receives click information and communication information, and combines the retrieval module to carry out patent push analysis. The specific push analysis process is as follows:

S11: Obtain patents that match the keywords and technical fields in the search information and mark them as primary patents;

S12: Mark the daily number of clicks on the primary patent within 10 days before the current system time as Bk, the viewing time of each click as Tki, the daily number of comments as Ck, and the daily number of reposts as Dk, The number of favorites per day is marked as Ek, and the number of likes per day is marked as Fk; k=1, 2, ..., 10; i=1, 2, ..., Bk;

S13: Mark the daily viewing time of the primary selection patent as Tk within 10 days before the current system time;

计算得出该初选专利每天的热度值Qk，其中，b1、b2、b3、r1、r2、r3和r4均为系数因子；S14: Utilize formulas

Calculate the daily heat value Qk of the primary patent, where b1, b2, b3, r1, r2, r3 and r4 are coefficient factors;

S15: Calculate the average heat value L of the primary patent in the 10 days before the current time according to the average calculation formula; obtain the standard deviation α of the daily heat value of the primary patent within the previous 10 days according to the standard deviation calculation formula, using the formula β=( L×η1-α×η2) ( ^η3+η4 ) calculates the continuous heat value β of the primary patent, wherein η1, η2, η3 and η4 are all coefficient factors;

S16: Mark the service evaluation coefficient as Ko, and calculate the average value of the service evaluation coefficient Ko to obtain the service evaluation mean value K;

S17: Mark the response time of the primary patent inventor to answer the user's question as J3o, where J3o=J2o-J1o, o=1, . J;

S18: Mark the total number of patents under the name of the primary patent inventor as P1; mark the number of patents that have been traded under the name of the primary patent inventor as P2;

计算得出该初选专利发明人的信誉值R，其中c1、c2、c3和c4均为系数因子；S19: Utilize formulas

Calculate the reputation value R of the primary patent inventor, where c1, c2, c3 and c4 are all coefficient factors;

得出该初选专利的推送值TS；其中d1、d2、d3、d4和d5为预设比例系数；λ＝0.00564327；P(x)为用户对该初选专利的兴趣值；S20: Utilize formulas

Get the push value TS of the primary patent; where d1, d2, d3, d4 and d5 are preset proportional coefficients; λ=0.00564327; P(x) is the user's interest in the primary patent;

The data analysis module transmits the push value TS to the server, and the server sorts the preliminary selection patents in descending order according to the push value TS and sends the sorted preliminary selection patents to the user terminal.

Further, the browsing module is used for the user terminal to browse patent information, and send the browsing record to the server; the browsing record includes browsing time, duration, and behavioral characteristics of comments, forwarding, favorites, and likes; the browsing time The time when the user clicks the patent link; the user behavior analysis module is used to receive and analyze the browsing records transmitted by the server; the specific steps include:

S41: Obtain the browsing time in the browsing record and mark the browsing time as Hx, the duration as Rx, the comment behavior value as S(C), the forwarding behavior value as S(D), and the favorite behavior value as S( E), the like behavior value is marked as S(F);

计算得出该条记录的时效值f(x)；其中g1为系数因子，Hx与TV越接近，则f(x)值越大；σ为预设因子；S42: Obtain the current time of the system, mark the current time as TV, and use the formula

Calculate the aging value f(x) of the record; where g1 is the coefficient factor, the closer Hx is to TV, the larger the f(x) value; σ is the preset factor;

S43: If the user has commented on the patent, then S(C)=1, otherwise S(C)=0; if the user has forwarded the patent, then S(D)=1, otherwise S(D)=0; If the user has a favorite of the patent, then S(E)=1, otherwise S(E)=0, if the user likes the patent, then S(F)=1, otherwise S(F)=0;

计算得出用户对该专利的兴趣值P(x)；其中g2为预设系数因子。S44: Utilize formulas

The user's interest value P(x) for the patent is calculated; where g2 is a preset coefficient factor.

Further, when the server receives the retrieval information transmitted by the retrieval module, it will automatically drive the control timing module to start timing, and when the server returns the retrieval result to the user terminal, it will transmit the retrieval signal to the server through the retrieval module, and will pass the retrieval signal when the browser is closed. The retrieval module transmits the solution signal to the server; the server drives the timing module to record the retrieval time and the solution time when receiving the response signal and the solution signal; the server marks the retrieval time as RT1 and transmits it to the evaluation module, the server Mark the resolution time as RT2 and transfer it to the evaluation module;

The evaluation module is used for the user to evaluate the patent retrieval service, and the evaluation rules are: score the retrieval service with a full score of 100 points; the specific working steps of the evaluation module are as follows:

S51: Mark the service score as Qs; obtain the number of patents browsed by the user during the entire retrieval process and mark it as Cs;

S52: Allocate weights according to the user's importance to the service score Qs, the number of browsed patents Cs, the retrieval time RT1 and the resolution time RT2 in the big data;

D1 is assigned to the service score Qs; D2 is assigned to the number of browsed patents Cs; D3 is assigned to the retrieval time RT1, and D4 is assigned to the solution time RT2; and D1+D2+D3+D4=1; D1> D2>D3>D4;

计算得到用户的检索满意值QR。S53: Utilize formulas

The user's retrieval satisfaction value QR is calculated.

Further, the evaluation module is used to transmit the retrieval satisfaction value QR to the server, and the server is used to stamp the retrieval satisfaction value QR with a timestamp and store it in the storage module and transmit the retrieval satisfaction value QR to the display module for real-time display; The basic information of the patent includes the patent including the inventor, the type of invention, the technical field and the name; the click information includes the number of clicks, the viewing time of each click, and the behavioral characteristics of comments, forwarding, favorites and likes; the communication information It includes the time J1x for the user to ask the question, the time for the inventor to answer the question J2x, the service evaluation coefficient, the total number of patents under the inventor's name, and the number of patents that have been traded under the inventor's name. The full score is 100 points.

Further, the intellectual property information retrieval method based on big data includes the following steps:

Step 1: The user registers and logs in through several user terminals, browses and checks the patent, and then publishes the retrieval information;

Step 2: The data analysis module receives click information and communication information, and carries out patent push analysis in combination with retrieval information; including:

X11: Obtain patents that match the keywords and technical fields in the search information and mark them as primary patents;

X12: Mark the daily number of clicks on the primary patent within 10 days before the current system time as Bk, the viewing time of each click as Tki, the daily number of comments as Ck, and the daily number of reposts as Dk, The number of favorites per day is marked as Ek, and the number of likes per day is marked as Fk;

X13: Mark the daily viewing time of the primary patent as Tk within 10 days before the current system time;

计算得出该初选专利每天的热度值Qk；X14: Utilize formulas

Calculate the daily heat value Qk of the primary patent;

X15: Calculate the average heat value L of the primary patent within 10 days before the current time according to the average calculation formula; obtain the standard deviation α of the daily heat value of the primary patent within the first 10 days according to the standard deviation calculation formula, using the formula β=( L×η1-α×η2) ^(η3+η4) is calculated to obtain the continuous heat value β of the primary patent;

X16: Mark the service evaluation coefficient as Ko, and calculate the average value of the service evaluation coefficient Ko to obtain the service evaluation mean value K;

X17: Mark the response time of the primary patent inventor to answer user questions as J3o; J3o=J2o-J1o; sum the response times J3o and take the average to obtain the average response time J;

X18: Mark the total number of patents under the name of the primary patent inventor as P1; mark the number of patents that have been traded under the name of the primary patent inventor as P2;

计算得出该初选专利发明人的信誉值R；X19: Utilize formulas

Calculate the reputation value R of the primary patent inventor;

得出该初选专利的推送值TS；X20: Utilize formula

Get the push value TS of the primary patent;

Step 3: Arrange the patents in descending order according to the push value TS and send the arranged patents to the user terminal;

Step 4: the user terminal browses the patent information through the browsing module, and sends the browsing record to the server; the user behavior analysis module receives the browsing record transmitted by the server and analyzes it; obtains the user's interest value P(x) in the patent; the specific steps are as follows:

X31: Get the browsing time in the browsing record and mark the browsing time as Hx, the duration as Rx, the comment behavior value as S(C), the forwarding behavior value as S(D), and the favorite behavior value as S (E), the like behavior value is marked as S(F);

计算得出该条记录的时效值f(x)；其中g1为系数因子，Hx与TV越接近，则f(x)值越大；σ为预设因子；X32: Get the current time of the system, mark the current time as TV, and use the formula

Calculate the aging value f(x) of the record; where g1 is the coefficient factor, the closer Hx is to TV, the larger the f(x) value; σ is the preset factor;

X33: If the user has commented on the patent, then S(C)=1, otherwise S(C)=0; if the user has forwarded the patent, then S(D)=1, otherwise S(D)=0; If the user has a favorite of the patent, then S(E)=1, otherwise S(E)=0, if the user likes the patent, then S(F)=1, otherwise S(F)=0;

计算得出用户对该专利的兴趣值P(x)；X34: Utilize formulas

Calculate the user's interest value P(x) for the patent;

Step 5: After the retrieval is completed, the user evaluates the patent retrieval service through the evaluation module, including:

X41: Mark the service score as Qs; get the number of patents viewed by the user throughout the search process and mark it as Cs;

X42: Allocate weights according to the user's importance to the service rating Qs, the number of browsed patents Cs, the retrieval time RT1 and the resolution time RT2 in the big data;

D1 is assigned to the service score Qs; D2 is assigned to the number of browsed patents Cs; D3 is assigned to the retrieval time RT1, and D4 is assigned to the solution time RT2; and D1+D2+D3+D4=1; D1> D2>D3>D4;

计算得到用户的检索满意值QR；X43: Utilize formulas

Calculate the user's retrieval satisfaction value QR;

Step 6: The server stamps the retrieval satisfaction value QR with a timestamp and stores it in the storage module, and transmits the retrieval satisfaction value QR to the display module for real-time display.

The beneficial effects of the present invention are:

得出该初选专利的推送值TS；服务器根据推送值TS对初选专利做降序排列并将排列后的初选专利发送至用户终端；巧妙利用大数据智能分析，提高检索效率；(1) In the present invention, the click information of each patent in the statistical database in the 10 days before the current time of the system is accessed by accessing the statistical module, and the communication information between the inventor and the user of the patent at every preset time is monitored by the investigation module; the data analysis module receives clicks Information and exchange information and combined with the retrieval module to carry out patent push analysis; firstly obtain patents that match the keywords and technical fields in the retrieval information and mark them as primary selection patents; combine relevant algorithms to obtain the continuous popularity value β of the primary selection patent; At the same time, the reputation value R of the primary patent inventor is obtained according to the exchange information; using the formula

The push value TS of the primary selection patent is obtained; the server sorts the primary selection patents in descending order according to the push value TS and sends the sorted primary selection patents to the user terminal; cleverly uses big data intelligent analysis to improve retrieval efficiency;

计算得出用户对该专利的兴趣值P(x)；结合持续热度值β和发明人的信誉值R，利用公式

得出该初选专利的推送值TS；使推送结果更准确，提高检索效率；(2) The present invention browses the patent information through the browsing module, and sends the browsing record to the server; the user behavior analysis module is used to receive the browsing record transmitted by the server and make an analysis; obtain the browsing time in the browsing record and mark the browsing time as Hx, Mark the duration as Rx, the comment behavior value as S(C), the forwarding behavior value as S(D), the favorite behavior value as S(E), and the like behavior value as S(F); using the formula

Calculate the user's interest value P(x) for the patent; combine the continuous popularity value β and the inventor's reputation value R, use the formula

Get the push value TS of the primary selection patent; make the push result more accurate and improve the retrieval efficiency;

计算得到用户的检索满意值QR；评价模块将检索满意值QR传输到服务器，服务器将检索满意值QR打上时间戳存储到存储模块并将检索满意值QR传输到显示模块进行实时显示，本发明对检索服务系统形成一个有效评价，方便后来查看。(3) The present invention evaluates the patent retrieval service through the evaluation module; assigns weights according to the importance of the user's service score Qs, the number of browsed patents Cs, the retrieval time RT1 and the resolution time RT2 in the big data; using the formula

The retrieval satisfaction value QR of the user is obtained by calculating; the evaluation module transmits the retrieval satisfaction value QR to the server, and the server stamps the retrieval satisfaction value QR with a timestamp and stores it in the storage module, and transmits the retrieval satisfaction value QR to the display module for real-time display. The retrieval service system forms an effective evaluation, which is convenient for later viewing.

Description of drawings

In order to facilitate the understanding of those skilled in the art, the present invention will be further described below with reference to the accompanying drawings.

FIG. 1 is a system block diagram of the present invention.

Detailed ways

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

As shown in Figure 1, the intellectual property information retrieval software management system and method based on big data includes a data acquisition module, several user terminals, a browsing module, a server, a retrieval module, a data acquisition module, a database and an evaluation module;

The user terminal is used to input the user's login information and registration information. The user logs in after entering the login information through the user terminal when there is an existing account. When the user does not have an account, the user enters the registration information through the user terminal to register a new account and log in for the first time;

The retrieval module is used for the user terminal to issue retrieval information and send the retrieval information to the server, and the retrieval information includes keywords and technical fields;

The data collection module is used to collect the basic information of each patent and transmit the basic information of each patent to the database for storage; the basic information of the patent includes the patent including the inventor, the type of invention, the technical field and the name; the database is used for the storage server to receive browsing records, evaluation records, retrieval information, login information and registration information;

The access statistics module is used to count the click information of each patent in the database within 10 days before the current time of the system and transmit the click information to the data analysis module; the click information includes the number of clicks, the viewing time of each click, comments, forwarding, favorites and The behavioral characteristics of likes; the investigation module is used to monitor the communication information between the inventor and the user of the patent every preset time and transmit the communication information to the data analysis module; the communication information includes the time when the user asks the question J1x, the time when the inventor answers the question. Time J2x, service evaluation coefficient, the total number of patents under the inventor's name, and the number of patents that have been traded under the inventor's name, the service evaluation coefficient rule is: score the inventor's service, the full score is 100 points;

The data analysis module receives the click information and communication information, and combines the retrieval module for patent push analysis. The specific push analysis process is as follows:

S11: Obtain patents that match the keywords and technical fields in the search information and mark them as primary patents;

S12: Mark the daily number of clicks on the primary patent within 10 days before the current system time as Bk, the viewing time of each click as Tki, the daily number of comments as Ck, and the daily number of reposts as Dk, The number of favorites per day is marked as Ek, and the number of likes per day is marked as Fk; k=1, 2, ..., 10; i=1, 2, ..., Bk;

S13: Mark the daily viewing time of the primary selection patent as Tk within 10 days before the current system time;

计算得出该初选专利每天的热度值Qk，其中，b1、b2、b3、r1、r2、r3和r4均为系数因子；S14: Utilize formulas

Calculate the daily heat value Qk of the primary patent, where b1, b2, b3, r1, r2, r3 and r4 are coefficient factors;

S15: Calculate the average heat value L of the primary patent in the 10 days before the current time according to the average calculation formula; obtain the standard deviation α of the daily heat value of the primary patent within the previous 10 days according to the standard deviation calculation formula, using the formula β=( L×η1-α×η2) ^(η3+η4) is calculated to obtain the continuous heat value β of the primary patent, wherein η1, η2, η3 and η4 are all coefficient factors;

S16: Mark the service evaluation coefficient as Ko, and calculate the average value of the service evaluation coefficient Ko to obtain the service evaluation mean value K;

S17: Mark the response time of the primary patent inventor to answer the user's question as J3o, J3o=J2o-J1o, o=1,...,n, sum the reaction times J3o and take the average to obtain the average reaction time J;

S18: Mark the total number of patents under the name of the primary patent inventor as P1; mark the number of patents that have been traded under the name of the primary patent inventor as P2;

计算得出该初选专利发明人的信誉值R，其中c1、c2、c3和c4均为系数因子；S19: Utilize formulas

Calculate the reputation value R of the primary patent inventor, where c1, c2, c3 and c4 are all coefficient factors;

得出该初选专利的推送值TS；其中d1、d2、d3、d4和d5为预设比例系数；λ＝0.00564327；P(x)为用户对该初选专利的兴趣值；S20: Utilize formulas

Get the push value TS of the primary patent; where d1, d2, d3, d4 and d5 are preset proportional coefficients; λ=0.00564327; P(x) is the user's interest in the primary patent;

The data analysis module transmits the push value TS to the server, and the server sorts the preliminary selection patents in descending order according to the push value TS and sends the sorted preliminary selection patents to the user terminal;

The browsing module is used by the user terminal to browse patent information and send the browsing record to the server; the browsing record includes the browsing time, duration, and behavioral characteristics of comments, forwarding, favorites and likes; the browsing time is the time when the user clicks the patent link; The user behavior analysis module is used to receive and analyze the browsing records transmitted by the server; the specific steps include:

S41: Obtain the browsing time in the browsing record and mark the browsing time as Hx, the duration as Rx, the comment behavior value as S(C), the forwarding behavior value as S(D), and the favorite behavior value as S( E), the like behavior value is marked as S(F);

计算得出该条记录的时效值f(x)；其中g1为系数因子，Hx与TV越接近，则f(x)值越大；σ为预设因子；S42: Obtain the current time of the system, mark the current time as TV, and use the formula

Calculate the aging value f(x) of the record; where g1 is the coefficient factor, the closer Hx is to TV, the larger the f(x) value; σ is the preset factor;

S43: If the user has commented on the patent, then S(C)=1, otherwise S(C)=0; if the user has forwarded the patent, then S(D)=1, otherwise S(D)=0; If the user has a favorite of the patent, then S(E)=1, otherwise S(E)=0, if the user likes the patent, then S(F)=1, otherwise S(F)=0;

计算得出用户对该专利的兴趣值P(x)；其中g2为预设系数因子；S44: Utilize formulas

Calculate the user's interest value P(x) for the patent; where g2 is a preset coefficient factor;

When the server receives the retrieval information transmitted by the retrieval module, it will automatically drive and control the timing module to start timing. When the server returns the retrieval result to the user terminal, it will transmit the retrieval signal to the server through the retrieval module. When the browser is closed, it will transmit the retrieval signal to the server through the retrieval module. Resolution signal; the server will drive the timing module to record the retrieval time and resolution time when it receives the response signal and resolution signal; the server marks the retrieval time as RT1 and transmits it to the evaluation module, and the server marks the resolution time as RT2 and sends it to the evaluation module transfer to the evaluation module;

The evaluation module is used for users to evaluate the patent retrieval service. The evaluation rules are: to score the retrieval service, the full score is 100 points; the specific working steps of the evaluation module are as follows:

S51: Mark the service score as Qs; obtain the number of patents browsed by the user during the entire retrieval process and mark it as Cs;

S52: Allocate weights according to the user's importance to the service score Qs, the number of browsed patents Cs, the retrieval time RT1 and the resolution time RT2 in the big data;

D1 is assigned to the service score Qs; D2 is assigned to the number of browsed patents Cs; D3 is assigned to the retrieval time RT1, and D4 is assigned to the solution time RT2; and D1+D2+D3+D4=1; D1> D2>D3>D4;

计算得到用户的检索满意值QR；S53: Utilize formulas

Calculate the user's retrieval satisfaction value QR;

The evaluation module is used to transmit the retrieval satisfaction value QR to the server, and the server is used to stamp the retrieval satisfaction value QR and store it in the storage module and transmit the retrieval satisfaction value QR to the display module for real-time display.

利用公式

计算得出该初选专利每天的热度值Qk；利用公式β＝(L×η1-α×η2)(^η3+η4)计算得出该初选专利的持续热度值β；The intellectual property information retrieval software management system and method based on big data, when working, users register and log in through several user terminals, browse and view patents through the browsing module, and then publish retrieval information through the retrieval module; the retrieval information includes: Keywords and technical fields; at the same time, the data collection module collects the basic information of each patent and transmits the basic information of each patent to the database for storage; accesses the statistics module to count the click information of each patent in the database within 10 days before the current system time The click information is transmitted to the data analysis module; the investigation module monitors the communication information between the inventor and the user of the patent every preset time and transmits the communication information to the data analysis module; the data analysis module receives the click information and communication information and combines with the retrieval module Carry out patent push analysis; first obtain patents that match the keywords and technical fields in the search information and mark them as primary patents; mark the number of clicks on the primary patents per day within 10 days before the current system time as Bk, and each time The clicked viewing time is marked as Tki, the daily number of comments is marked as Ck, the daily number of reposts is marked as Dk, the daily number of favorites is marked as Ek, and the daily number of likes is marked as Fk;

Use the formula

Calculate the daily heat value Qk of the primary patent; use the formula β=(L×η1-α×η2)( ^η3+η4 ) to calculate the continuous heat value β of the primary patent;

计算得出该初选专利发明人的信誉值R；利用公式

得出该初选专利的推送值TS；服务器根据推送值TS对初选专利做降序排列并将排列后的初选专利发送至用户终端；Mark the service evaluation coefficient as Ko, take the average value of the service evaluation coefficient Ko to obtain the service evaluation mean value K, mark the response time of the primary patent inventor to answer the user's question as J3o, sum the response time J3o and take the average value to get Average response time J; mark the total number of patents under the name of the inventor of the primary patent as P1; mark the number of patents that have been traded under the name of the inventor of the primary patent as P2; use the formula

Calculate the reputation value R of the primary patent inventor; use the formula

Obtain the push value TS of the primary selection patent; the server sorts the primary selection patents in descending order according to the push value TS and sends the sorted primary selection patents to the user terminal;

计算得出该条记录的时效值f(x)；The user terminal browses the patent information through the browsing module, and sends the browsing record to the server; the user behavior analysis module is used to receive the browsing record transmitted by the server and analyze it; obtain the browsing time in the browsing record and mark the browsing time as Hx, which will last for a long time. It is marked as Rx, the comment behavior value is marked as S(C), the forwarding behavior value is marked as S(D), the favorite behavior value is marked as S(E), and the like behavior value is marked as S(F); to obtain the current time of the system, Mark the current time as TV, using the formula

Calculate the aging value f(x) of the record;

计算得出用户对该专利的兴趣值P(x)；Use the formula

Calculate the user's interest value P(x) for the patent;

计算得到用户的检索满意值QR；评价模块将检索满意值QR传输到服务器，服务器将检索满意值QR打上时间戳存储到存储模块并将检索满意值QR传输到显示模块进行实时显示。The evaluation module is used for users to evaluate the patent retrieval service. The evaluation rules are: score the retrieval service with a full score of 100; first mark the service score as Qs; obtain the number of patents browsed by the user during the entire retrieval process and mark it as Cs, assign weights according to the importance of users in the big data to the service score Qs, the number of browsed patents Cs, the retrieval time RT1 and the resolution time RT2; using the formula

The user's retrieval satisfaction value QR is calculated; the evaluation module transmits the retrieval satisfaction value QR to the server, and the server stamps the retrieval satisfaction value QR with a timestamp and stores it in the storage module, and transmits the retrieval satisfaction value QR to the display module for real-time display.

The above formulas are obtained by collecting a large amount of data for software simulation and corresponding experts for parameter setting processing, and obtaining formulas that are consistent with the real results.

The above-disclosed preferred embodiments of the present invention are provided only to help illustrate the present invention. The preferred embodiments do not describe all the details and do not limit the invention to specific embodiments only. Obviously, many modifications and variations are possible in light of the content of this specification. These embodiments are selected and described in this specification in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can well understand and utilize the present invention. The present invention is to be limited only by the claims and their full scope and equivalents.

Claims (5)

1. The intellectual property information retrieval software management system based on big data is characterized by comprising a data acquisition module, a plurality of user terminals, a browsing module, a user behavior analysis module, a server, a retrieval module, a data acquisition module, a database and an evaluation module;

the user terminal is used for inputting login information and registration information of a user, the user logs in after inputting the login information through the user terminal when the user has an account, and the user logs in for the first time after inputting the registration information through the user terminal to register a new account when the user does not have the account;

the retrieval module is used for the user terminal to issue retrieval information and send the retrieval information to the server, and the retrieval information comprises keywords and the technical field;

the data acquisition module is used for acquiring the basic information of each patent and transmitting the basic information of each patent to the database for storage; the database is used for storing browsing records, evaluation records, retrieval information, login information and registration information received by the server;

the access statistical module is used for counting click information of each patent in the database within 10 days before the current time of the system and transmitting the click information to the data analysis module; the investigation module is used for monitoring the communication information between the inventor of the patent and the user at preset time intervals and transmitting the communication information to the data analysis module;

the data analysis module receives click information and exchange information and performs pushing analysis of patents by combining with the retrieval module, and the specific pushing analysis process is as follows:

s11: acquiring patents which accord with keywords and technical fields in the retrieval information and marking the patents as primary-selected patents;

s12: marking the clicked times of the primary selection patent every day within 10 days before the current time of the system as Bk, marking the watching time of each click as Tki, marking the commented times every day as Ck, marking the forwarded times every day as Dk, marking the collected times every day as Ek, and marking the praised times every day as Fk; k is 1, 2, …, 10; i ═ 1, 2, …, Bk;

s13: marking the time when the primary selected patent is watched every day within 10 days before the current time of the system as Tk;

s14: using formulas

Calculating to obtain a heat value Qk of the initially selected patent every day, wherein b1, b2, b3, r1, r2, r3 and r4 are coefficient factors;

s15: obtaining an average heat value L of the initially selected patent within 10 days before the current time according to an average value calculation formula; obtaining the standard deviation alpha of the calorific value of the initially selected patent every day in the first 10 days according to a standard deviation calculation formula, and utilizing the formula

Calculating to obtain a continuous heat value beta of the initially selected patent, wherein eta 1, eta 2, eta 3 and eta 4 are coefficient factors;

s16: marking the service evaluation coefficient as Ko, and calculating the average value of the service evaluation coefficient Ko to obtain a service evaluation mean value K;

s17: marking the response time of the initially selected patent inventor for answering the user question as J3o, wherein J3o is J2o-J1o, o is 1, and n, and summing and averaging the response times J3o to obtain an average response time J;

s18: marking the total number of patents in the name of the initially selected patent inventor as P1; marking the number of patents which have been committed in the name of the initially selected patent inventor as P2;

s19: using formulas

Calculating to obtain a reputation value R of the initially selected patent inventor, wherein c1, c2, c3 and c4 are coefficient factors;

s20: using formulas

Obtaining a pushing value TS of the initially selected patent; wherein d1, d2, d3, d4 and d5 are preset proportionality coefficients; λ 0.00564327; p (x) is the interest value of the user in the initially selected patent;

and the data analysis module transmits the pushing value TS to the server, and the server performs descending arrangement on the primary selected patents according to the pushing value TS and sends the arranged primary selected patents to the user terminal.

2. The intellectual property information retrieval software management system based on big data as claimed in claim 1, wherein the browsing module is used for the user terminal to browse the patent information and send the browsing record to the server; the browsing record comprises browsing time, duration and behavior characteristics of comment, forwarding, collection and praise; the browsing time is the time for the user to click on the patent link; the user behavior analysis module is used for receiving the browsing record transmitted by the server and analyzing the browsing record; the method comprises the following specific steps:

s41: acquiring browsing time in the browsing record, marking the browsing time as Hx, marking the duration as Rx, marking the comment behavior value as S (C), marking the forwarding behavior value as S (D), marking the collection behavior value as S (E), and marking the praise behavior value as S (F);

s42: obtaining the current time of the system, marking the current time as TV, and utilizing a formula

Calculating to obtain the aging value f (x) of the record; where g1 is a coefficient factor, the closer Hx is to TV, the greater the value of f (x); sigma is a preset factor;

s43: if the user has a comment on the patent, s (c) is 1, otherwise s (c) is 0; if the user has forwarding to the patent, s (d) is 1, otherwise s (d) is 0; if the user has a collection for the patent, s (e) ═ 1, otherwise s (e) ═ 0, if the user has a approval for the patent, s (f) ═ 1, otherwise s (f) ═ 0;

s44: using formulas

Calculating to obtain a user interest value P (x) of the patent; where g2 is a preset coefficient factor.

3. The intellectual property information retrieval software management system based on big data as claimed in claim 1, wherein the server automatically drives and controls the timing module to start timing when receiving the retrieval information transmitted by the retrieval module, transmits the retrieval signal to the server through the retrieval module when the server returns the retrieval result to the user terminal, and transmits the solution signal to the server through the retrieval module when the browser is closed; the server drives the timing module to record retrieval time and solution time when receiving the response signal and the solution signal; the server marks the retrieval time as RT1 and transmits it to the evaluation module, the server marks the resolution time as RT2 and transmits it to the evaluation module;

the evaluation module is used for evaluating the retrieval service of the patent by the user, and the evaluation rule is as follows: scoring the retrieval service, wherein the full score is 100; the specific working steps of the evaluation module are as follows:

s51: marking the service score as Qs; acquiring the number of the patents browsed by the user in the whole retrieval process and marking the patents as Cs;

s52: assigning weights according to the degree of importance of the users to the service scores Qs, the number Cs of browsed patents, the retrieval time RT1 and the resolution time RT2 in the big data;

assigning a weight of D1 to the service score Qs; assigning a weight D2 to the number Cs of viewed patents; the retrieval time RT1 is assigned with the weight D3, and the solution time RT2 is assigned with the weight D4; and D1+ D2+ D3+ D4 is 1; d1> D2> D3> D4;

s53: using formulas

And calculating to obtain a retrieval satisfaction value QR of the user.

4. The intellectual property information retrieval software management system based on big data as claimed in claim 1, wherein the evaluation module is used for transmitting the retrieval satisfaction value QR to the server, the server is used for storing the retrieval satisfaction value QR to the storage module by stamping a time stamp and transmitting the retrieval satisfaction value QR to the display module for real-time display; the basic information of the patent includes the patent including the inventor, the type of invention, the technical field and the name; the click information comprises click times, the watching time of each click and behavior characteristics of comment, forwarding, collection and praise; the communication information comprises time J1x of a user proposing a question, time J2x of an inventor answering the question, service evaluation coefficient, total number of patents under the name of the inventor and number of patents which have been submitted under the name of the inventor, and the service evaluation coefficient rules are as follows: the inventor was given a service score of 100 points.

5. The intellectual property information retrieval method based on the big data is characterized by comprising the following steps:

the method comprises the following steps: a user registers and logs in through a plurality of user terminals, browses and checks patents, and then releases retrieval information;

step two: the data analysis module receives click information and exchange information and performs pushing analysis on patents by combining retrieval information; the method comprises the following steps:

x11: acquiring patents which accord with keywords and technical fields in the retrieval information and marking the patents as primary-selected patents;

x12: marking the clicked times of the primary selection patent every day within 10 days before the current time of the system as Bk, marking the watching time of each click as Tki, marking the commented times every day as Ck, marking the forwarded times every day as Dk, marking the collected times every day as Ek, and marking the praised times every day as Fk;

x13: marking the time when the primary selected patent is watched every day within 10 days before the current time of the system as Tk;

x14: using formulas

Calculating to obtain a heat value Qk of the initially selected patent every day;

x15: obtaining an average heat value L of the initially selected patent within 10 days before the current time according to an average value calculation formula; obtaining the standard deviation alpha of the calorific value of the initially selected patent every day in the first 10 days according to a standard deviation calculation formulaBy the formula

Calculating to obtain a continuous heat value beta of the initially selected patent;

x16: marking the service evaluation coefficient as Ko, and calculating the average value of the service evaluation coefficient Ko to obtain a service evaluation mean value K;

x17: the reaction time of the primary selected patent inventor for answering the user question is marked as J3 o; j3o ═ J2o-J1 o; summing the reaction times J3o and averaging to obtain an average reaction time J;

x18: marking the total number of patents in the name of the initially selected patent inventor as P1; marking the number of patents which have been committed in the name of the initially selected patent inventor as P2;

x19: using formulas

Calculating to obtain a reputation value R of the initially selected patent inventor;

x20: using formulas

Obtaining a pushing value TS of the initially selected patent;

step three: sorting the patents in a descending order according to the TS and sending the sorted patents to the user terminal;

step four: the user terminal browses the patent information through the browsing module and sends a browsing record to the server; the user behavior analysis module receives the browsing record transmitted by the server and analyzes the browsing record; obtaining the interest value P (x) of the user to the patent; the method comprises the following specific steps:

x31: acquiring browsing time in the browsing record, marking the browsing time as Hx, marking the duration as Rx, marking the comment behavior value as S (C), marking the forwarding behavior value as S (D), marking the collection behavior value as S (E), and marking the praise behavior value as S (F);

x32: obtaining the current time of the system, marking the current time as TV, and utilizing a formula

Calculating to obtain the aging value f (x) of the record; where g1 is a coefficient factor, the closer Hx is to TV, the greater the value of f (x); sigma is a preset factor;

x33: if the user has a comment on the patent, s (c) is 1, otherwise s (c) is 0; if the user has forwarding to the patent, s (d) is 1, otherwise s (d) is 0; if the user has a collection for the patent, s (e) ═ 1, otherwise s (e) ═ 0, if the user has a approval for the patent, s (f) ═ 1, otherwise s (f) ═ 0;

x34: using formulas

Calculating to obtain a user interest value P (x) of the patent;

step five: after the retrieval is completed, the user evaluates the retrieval service of the patent through an evaluation module, which comprises the following steps:

x41: marking the service score as Qs; acquiring the number of the patents browsed by the user in the whole retrieval process and marking the patents as Cs;

x42: assigning weights according to the degree of importance of the users to the service scores Qs, the number Cs of browsed patents, the retrieval time RT1 and the resolution time RT2 in the big data;

assigning a weight of D1 to the service score Qs; assigning a weight D2 to the number Cs of viewed patents; the retrieval time RT1 is assigned with the weight D3, and the solution time RT2 is assigned with the weight D4; and D1+ D2+ D3+ D4 is 1; d1> D2> D3> D4;

x43: using formulas

Calculating to obtain a retrieval satisfaction value QR of the user;

step six: and the server stamps a time stamp on the retrieval satisfaction value QR and stores the retrieval satisfaction value QR in a storage module and transmits the retrieval satisfaction value QR to a display module for real-time display.

Images