CN105653714A - Knowledge pushing method based on intelligent capturing - Google Patents

Abstract

The invention relates to a knowledge pushing method based on intelligent capturing. The knowledge pushing method comprises the following steps that 1, user information is acquired, and keywords are determined; 2, a keyword table is pushed according to keyword retrieval, if the keywords exist, the step 3 is performed for knowledge object querying and matching, and if not, association is built for the keywords and all usable knowledge objects, and then the step 3 is performed; 3, the knowledge objects are queried according to the keywords, and multiple knowledge objects are screened according to the matching degree of the keywords and the knowledge objects from high to low; 4, user conditions are acquired, and the matching degree of the knowledge objects and a user is calculated according to the user conditions; 5, the knowledge objects meeting the requirement are screened out according to the matching degree; 6, the screened knowledge objects are pushed to the user and sorted according to the matching degree from high to low. According to the knowledge pushing method, traditional knowledge searching is changed into automatic knowledge analyzing and pushing, and therefore the purpose that knowledge follows the procedures is achieved; precise and individual knowledge pushing can be achieved, and not only is the design efficiency improved, but also a knowledge asset system of an enterprise can be formed.

Description

A kind of knowledge method for pushing based on intelligent capture

Technical field

The invention belongs to knowledge acquisition technology field, it relates to a kind of knowledge method for pushing, is specifically related to a kind of knowledge method for pushing based on intelligent capture.

Background technology

Slip-stick artist is in the process of design research and development, it is necessary to use software different in a large number, database, document library etc. In design R&D process, involved many experiences all leave in different software architectures from knowledge, this just makes the work that slip-stick artist has to stop at hand in working process in order to obtain required knowledge, open different software, database and document library etc., go to look for relevant data, parameter, experience etc., thus waste a lot of time, design efficiency is very low.

In order to solve the problem, some system also employs knowledge propelling movement technology, to push technology to user's active push knowledge by knowledge. But, these knowledge pushes technology generally all cannot be different according to the role of user, and the difference of the use habit etc. of user and change the knowledge content of propelling movement. On the contrary, specific or similar knowledge can only be pushed to user. Like this, cause that the knowledge of propelling movement is inaccurate and amount is bigger etc. , in addition it is also necessary to the knowledge pushed is recognized and analyzes by user, in addition sometimes on the contrary in the time of waste slip-stick artist.

In view of the above-mentioned technological deficiency of prior art, urgently need a kind of novel knowledge method for pushing of development.

Summary of the invention

It is an object of the invention to overcome above-mentioned defect of the prior art, a kind of knowledge method for pushing based on intelligent capture is provided, it can complete the active transfer of knowledge experience, and can the difference of exploitation software residing for slip-stick artist and environment and push different knowledge, so that the knowledge accuracy height pushed, it is possible to reduce the time that slip-stick artist finds knowledge.

In order to realize above-mentioned purpose, the present invention provides following technical scheme: a kind of knowledge method for pushing based on intelligent capture, and it comprises the following steps:

1) obtain user profile and determine keyword according to described user profile;

2) crucial word table is pushed according to the retrieval of described keyword, if there is described keyword in the crucial word table of described propelling movement, so proceed to step 3), if there is not described keyword in described crucial word table, so it is associated for described keyword and all available Object of Knowledge, and then proceeds to step 3);

3) according to described keyword query Object of Knowledge, and sieve from high to low according to the matching degree of described keyword and Object of Knowledge and get many Object of Knowledge;

4) obtain user situation, and calculate the matching degree being sieved many Object of Knowledge and the user got according to described user situation;

5) Object of Knowledge meeting and pushing parameter request is filtered out according to the matching degree of described many Object of Knowledge and user;

6) by described step 5) Object of Knowledge pushing parameter request that meets that filters out is pushed to user and sorts from high to low according to the matching degree of described many Object of Knowledge and user.

Further, wherein, described step 1) in user profile comprise the post of user, technical ability, position and hobby.

Further, wherein, described step 4) in user situation comprise service condition in user grouping service condition, User IP grouping service condition, user's service condition and user's special time period.

Compared with existing knowledge method for pushing, the knowledge method for pushing based on intelligent capture of the present invention has following Advantageous Effects:

The knowledge method for pushing based on intelligent capture of the present invention have employed half machine intelligent mode, automatically pushes the knowledge closer to user for different personnel, it is possible to reach accurate, personalized propelling movement. In addition, it is engraved in during the knowledge method for pushing of the present invention and collects a large amount of user's daily records for automatically analyzing to push, comprise the click of user, browse, the habit such as search, and in conjunction with user profile such as post, responsibility etc., screen the most applicable user from knowledge base and reach the knowledge pushing valve value, and automatically it is displayed in user's environment, greatly improve the service efficiency of user. Like this, it is possible to help slip-stick artist to free from repeatability brainwork, energy is focused on research of technique and comes up with innovation, remarkable improving product R & D design ability.

Accompanying drawing explanation

Fig. 1 is the schema of the knowledge method for pushing based on intelligent capture of the present invention.

Embodiment

Below in conjunction with drawings and Examples, the present invention is further described, and the content of embodiment is not as the restriction to protection scope of the present invention.

The present invention relates to the knowledge method for pushing based on intelligent capture, its systematize being the operating mode to traditional product R & D design carries out improves, and can form the design environment new model of " knowledge is with flow process ". This can help slip-stick artist to free from repeatability brainwork, energy focuses on research of technique and comes up with innovation, remarkable improving product R & D design ability.

Fig. 1 shows the schema of the knowledge method for pushing based on intelligent capture of the present invention. As indicated with 1, the knowledge method for pushing based on intelligent capture of the present invention mainly comprises the following steps:

First, automatic capturing user profile and determine keyword according to described user profile. In the present invention, described user profile comprises the post of user, technical ability, position and hobby etc. Pass through user profile, it is possible to understand the role of user, and then be convenient to push according to the difference of role different knowledge. Such as, if the post of a user is design engineer, the post of another user is Simulation Engineering teacher, and so their role is different, it is determined that keyword different, the knowledge of propelling movement is naturally also different.

Secondly, push crucial word table according to the retrieval of described keyword, judge whether the crucial word table of described propelling movement exists described keyword. In the present invention, the crucial word table of described propelling movement stores propelling movement keyword. The only described keyword pushing existence in crucial word table, just can set up with Object of Knowledge described later and contact. If a keyword is not present in the crucial word table of described propelling movement, then can not go with this keyword to retrieve Object of Knowledge described later, it is necessary to first set up, by between this keyword with Object of Knowledge, the relation of association, inquiry Object of Knowledge could be gone with this keyword. Crucial word table in the present invention is the place for converging and store keyword, is obtained by the key attribute of definition time newly-increased Object of Knowledge (such as knowledge entry).

Wherein, if the crucial word table of described propelling movement exists described keyword, next step is so proceeded to. If the crucial word table of described propelling movement does not exist described keyword, so need for described keyword and all available Object of Knowledge are associated, and then proceed to next step. In the present invention, in the event of new keyword, such as, new post has there is, such as testing engineering teacher. And search for less than the keyword with testing engineering Shi Xiangguan in described crucial word table. Now, it is necessary to all Object of Knowledge used with testing engineering Shi Keneng with the keyword of testing engineering Shi Xiangguan are associated, and described keyword is stored in the crucial word table of described propelling movement, so that follow-up use.

Then, according to described keyword query Object of Knowledge, and sieve from high to low according to the matching degree of described keyword and Object of Knowledge and get many Object of Knowledge. In the present invention, it is included in the title of Object of Knowledge, keyword and content according to described keyword query Object of Knowledge and retrieves described keyword. Meanwhile, in the present invention, will calculating the matching degree of described keyword and Object of Knowledge, matching degree is more high, shows that this Object of Knowledge is more relevant to described keyword. Owing to the low especially Object of Knowledge of matching degree is not the knowledge needed for user usually, therefore, it is only necessary to sieve from high in the end according to matching degree get a certain amount of Object of Knowledge can (that is, matching degree not be sieved lower than the Object of Knowledge of certain value and is got).

In the present invention, it may be preferred that by the matching degree of following formulae discovery keyword and Object of Knowledge:

$O_m=\frac{T_m\×20+K_m\×30+C_m\×50}{100}\×100\%$

Variable declaration

O_m: the matching degree of keyword and Object of Knowledge;

T_m: the matching degree of the title of keyword and Object of Knowledge, span is 0��100; K_m: close

The matching degree of the keyword of key word and Object of Knowledge, span is 0��100;

C_m: the matching degree of the content of keyword and Object of Knowledge, span is 0��100.

The title of Object of Knowledge often occurring once, the value of the matching degree of the mark of described keyword, keyword and Object of Knowledge just adds one. With reason, the keyword of Object of Knowledge often occurring once, the value of the matching degree of the keyword of described keyword, keyword and Object of Knowledge just adds one. The content of Object of Knowledge often occurring once, the value of the matching degree of the content of described keyword, keyword and Object of Knowledge just adds one.

The matching degree between keyword and Object of Knowledge is gone out, and the height according to matching degree filters out many Object of Knowledge, for propelling movement by above-mentioned formulae discovery.

Then, automatic capturing user situation, and the matching degree being sieved many Object of Knowledge and the user got is calculated according to described user situation.

In the present invention, it may be preferred that described user situation comprises service condition in user grouping service condition, User IP grouping service condition, user's service condition and user's special time period.

In addition, in the present invention, it may be preferred that respectively by the matching degree of following formulae discovery Object of Knowledge and user.

The calculation formula of user grouping service condition is as follows:

${\mathrm{UGO}}_w=\frac{{\mathrm{UGC}}_o+{\mathrm{UGC}}_n}{{\mathrm{UGS}}_o+{\mathrm{UGS}}_n+{\mathrm{UGP}}_o+{\mathrm{UGP}}_n}\×100\%$

Variable declaration:

UGO_w: Object of Knowledge is by user grouping coupling weights; |

UGC_o: the number of clicks that Object of Knowledge has been added up by user grouping;

UGC_n: the number of clicks that Object of Knowledge is not added up by user grouping;

UGS_o: the searching times that Object of Knowledge has been added up by user grouping;

UGS_n: the searching times that Object of Knowledge is not added up by user grouping;

UGP_o: the propelling movement number of times that Object of Knowledge has been added up by user grouping;

UGP_n: Object of Knowledge is by the propelling movement number of times that user grouping is statistics.

The calculation formula of User IP grouping service condition is as follows:

${\mathrm{UIGO}}_w\frac{{\mathrm{UIGC}}_o+{\mathrm{UIGC}}_n}{{\mathrm{UIGS}}_o+{\mathrm{UIGS}}_n+{\mathrm{UIGP}}_o+{\mathrm{UIGP}}_n}\×100\%$

Variable declaration:

UIGO_w: Object of Knowledge presses User IP group match weights;

UIGC_��: the number of clicks that Object of Knowledge has been added up by User IP grouping;

UIGC_n: the number of clicks that Object of Knowledge is not added up by User IP grouping;

UIGS_o: the searching times that Object of Knowledge has been added up by User IP grouping;

UIGS_n: the searching times that Object of Knowledge is not added up by User IP grouping;

UIGP_o: the propelling movement number of times that Object of Knowledge has been added up by User IP grouping;

UIGP_n: the propelling movement number of times that Object of Knowledge is not added up by User IP grouping.

The calculation formula of user's service condition is as follows:

${\mathrm{UUO}}_w=\frac{{\mathrm{UUC}}_o+{\mathrm{UUC}}_n}{{\mathrm{UUS}}_o+{\mathrm{UUS}}_n+{\mathrm{UUP}}_o+{\mathrm{UUP}}_n}\×100\%$

Variable declaration:

UUO_w: Object of Knowledge mates weights by user;

UUC_o: the number of clicks that Object of Knowledge has been added up by user;

UUC_n: the number of clicks that Object of Knowledge is not added up by user;

UUS_o: the searching times that Object of Knowledge has been added up by user;

UUS_n: the searching times that Object of Knowledge is not added up by user;

UUP_o: the propelling movement number of times that Object of Knowledge has been added up by user;

UUP_n: the propelling movement number of times that Object of Knowledge is not added up by user.

In special time period, the calculation formula of service condition is as follows:

${\mathrm{TO}}_w=\frac{T_c}{T_s+T_p}\×100\%$

Variable declaration:

TO_w: the weights of Object of Knowledge in special time period;

T_c: the number of clicks of Object of Knowledge in special time period;

T_s: the searching times of Object of Knowledge in special time period;

T_p: the propelling movement number of times of Object of Knowledge in special time period.

The matching degree of various user's service condition can be calculated according to above-mentioned formula.

Then, so that it may filter out the Object of Knowledge meeting and pushing parameter request with the matching degree according to described many Object of Knowledge and user. Wherein, it is possible to for each matching degree sets a threshold value, only when matching degree is higher than this threshold value, this Object of Knowledge is just pushed; When matching degree is lower than high threshold, do not push this Object of Knowledge.

Finally, the Object of Knowledge meeting propelling movement parameter request filtered out is pushed to user and sorts from high to low according to the matching degree of described many Object of Knowledge and user, can for user.

Traditional scientific research personnel is searched knowledge and changes analysis propelling movement knowledge automatically into by the knowledge method for pushing based on intelligent capture of the present invention, and accurate, personalized propelling movement can be reached, thus not only increase substantially design efficiency, more can accumulate the Knowledge Assets system forming enterprise gradually, have broad application prospects.

The above embodiment of the present invention is only for example of the present invention is clearly described, and is not the restriction to embodiments of the present invention. For those of ordinary skill in the field, can also make other changes in different forms on the basis of the above description. Here cannot all enforcement modes be given exhaustive. The apparent change that the technical scheme of every the present invention of belonging to is extended out or variation are still in the row of protection scope of the present invention.

Claims (8)

1. the knowledge method for pushing based on intelligent capture, it is characterised in that: comprise the following steps

1) obtain user profile and determine keyword according to described user profile;

2) crucial word table is pushed according to the retrieval of described keyword, if there is described keyword in the crucial word table of described propelling movement, so proceed to step 3), if there is not described keyword in described crucial word table, so it is associated for described keyword and all available Object of Knowledge, and then proceeds to step 3);

3) according to described keyword query Object of Knowledge, and sieve from high to low according to the matching degree of described keyword and Object of Knowledge and get many Object of Knowledge;

4) obtain user situation, and calculate the matching degree being sieved many Object of Knowledge and the user got according to described user situation;

5) Object of Knowledge meeting and pushing parameter request is filtered out according to the matching degree of described many Object of Knowledge and user;

6) by described step 5) Object of Knowledge pushing parameter request that meets that filters out is pushed to user and sorts from high to low according to the matching degree of described many Object of Knowledge and user.

2. the knowledge method for pushing based on intelligent capture according to claim 1, it is characterised in that: described step 1) in user profile comprise the post of user, technical ability, position and hobby, conventional keyword.

3. the knowledge method for pushing based on intelligent capture according to claim 2, it is characterised in that: described step 4) in user situation comprise service condition in user grouping service condition, User IP grouping service condition, user's service condition and user's special time period.

4. the knowledge method for pushing based on intelligent capture according to claim 1 or 2 or 3, it is characterised in that:

By the matching degree of following formulae discovery keyword and Object of Knowledge:

$O_m=\frac{T_m\×20+K_m\×30+C_m\×50}{100}\×100\%.$

O_m: the matching degree of keyword and Object of Knowledge;

T_m: the matching degree of the title of keyword and Object of Knowledge, span is 0��100; K_m: the matching degree of the keyword of keyword and Object of Knowledge, span is 0��100;

C_m: the matching degree of the content of keyword and Object of Knowledge, span is 0��100;

The title of Object of Knowledge often occurring once, the value of the matching degree of the mark of described keyword, keyword and Object of Knowledge just adds one; With reason, the keyword of Object of Knowledge often occurring once, the value of the matching degree of the keyword of described keyword, keyword and Object of Knowledge just adds one; The content of Object of Knowledge often occurring once, the value of the matching degree of the content of described keyword, keyword and Object of Knowledge just adds one.

5. the knowledge method for pushing based on intelligent capture according to claim 4, it is characterised in that:

The calculation formula of user grouping service condition is as follows:

${\mathrm{UGO}}_w\frac{{\mathrm{UGC}}_o+{\mathrm{UGC}}_n}{{\mathrm{UGS}}_o+{\mathrm{UGS}}_n+{\mathrm{UGP}}_o+{\mathrm{UGP}}_n}\×100\%$

Variable declaration:

UGO_w: Object of Knowledge is by user grouping coupling weights;

UGC_o: the number of clicks that Object of Knowledge has been added up by user grouping;

UGC_n: the number of clicks that Object of Knowledge is not added up by user grouping;

UGS_o: the searching times that Object of Knowledge has been added up by user grouping;

UGS_n: the searching times that Object of Knowledge is not added up by user grouping;

UGP_o: the propelling movement number of times that Object of Knowledge has been added up by user grouping;

UGP_n: the propelling movement number of times that Object of Knowledge is not added up by user grouping.

6. the knowledge method for pushing based on intelligent capture according to claim 4, it is characterised in that:

The calculation formula of User IP grouping service condition is as follows:

${\mathrm{UIGO}}_w\frac{{\mathrm{UIGC}}_o+{\mathrm{UIGC}}_n}{{\mathrm{UIGS}}_o+{\mathrm{UIGS}}_n+{\mathrm{UIGP}}_o+{\mathrm{UIGP}}_n}\×100\%$

Wherein:

UIGO_w: Object of Knowledge presses User IP group match weights;

UIGC_o: the number of clicks that Object of Knowledge has been added up by User IP grouping;

UIGC_n: the number of clicks that Object of Knowledge is not added up by User IP grouping;

UIGS_o: the searching times that Object of Knowledge has been added up by User IP grouping;

UIGS_n: the searching times that Object of Knowledge is not added up by User IP grouping;

UIGP_o: the propelling movement number of times that Object of Knowledge has been added up by User IP grouping;

UIGP_n: the propelling movement number of times that Object of Knowledge is not added up by User IP grouping.

7. the knowledge method for pushing based on intelligent capture according to claim 4, it is characterised in that:

The calculation formula of user's service condition is as follows:

${\mathrm{UUO}}_w\frac{{\mathrm{UUC}}_o+{\mathrm{UUC}}_n}{{\mathrm{UUS}}_o+{\mathrm{UUS}}_n+{\mathrm{UUP}}_o+{\mathrm{UUP}}_n}\×100\%$

Variable declaration:

UUO_w: Object of Knowledge mates weights by user;

UUC_o: the number of clicks that Object of Knowledge has been added up by user;

UUC_n: the number of clicks that Object of Knowledge is not added up by user;

UUS_o: the searching times that Object of Knowledge has been added up by user;

UUS_n: the searching times that Object of Knowledge is not added up by user;

UUP_o: the propelling movement number of times that Object of Knowledge has been added up by user;

UUP_n: the propelling movement number of times that Object of Knowledge is not added up by user.

8. the knowledge method for pushing based on intelligent capture according to claim 4, it is characterised in that:

In special time period, the calculation formula of service condition is as follows:

${\mathrm{TO}}_w=\frac{T_c}{T_s+T_p}\×100\%$

Variable declaration:

TO_w: the weights of Object of Knowledge in special time period;

T_c: the number of clicks of Object of Knowledge in special time period;

T_s: the searching times of Object of Knowledge in special time period;

T_p: the propelling movement number of times of Object of Knowledge in special time period.