TW201131501A - Active knowledge managing system, method and computer program product for solving questions - Google Patents

Abstract

An active knowledge managing system for solving questions comprises a knowledge document database storing multiple knowledge documents and eigenvectors coded by work technology codes; an automatic replying module receiving a question request instruction and transforming the question into a question eigenvector, calculating the inner product between the question eigenvector and the eigenvector of the knowledge documents in the knowledge document database, and providing the knowledge documents with the inner product value higher than a default value to a user; and an automatic assigning module, when the knowledge documents with the inner product value higher than the default value do not exist, the automatic assigning module will search an expert that has ever dealt with a case corresponding to the work technology codes in the knowledge document database according to the work technology codes corresponding to the question, and notify the expert to reply.

Description

201131501 VI. Description of the Invention: [Technical Field] The present invention relates to a knowledge management system (KM), and more particularly to an active knowledge management system for problem solving. [Prior Art] Accumulation of experience is important in a business or organization; but if empirical knowledge accumulates only in certain individuals, 'distributed knowledge' is not the most effective and often difficult to fully inherit. Therefore, systematic knowledge management techniques are born. Existing knowledge management related technologies such as: 1. The Republic of China No. 359778 invention patent "Expert-assisted problem solving system" generally proposes that when the "Problem Solution Index Module" cannot answer the received question message, An "Expert Information Database" is connected to obtain expert personal data and professional domain communications. However, the focus of the case's appeal is that "users can use handwriting to propose __"' but do not use the algorithm (eg, "How does the problem solution index module provide answers?", "How to find the corresponding problem The experts?" is not in the ink, it should be just a general search method, the accuracy is limited. 2. The Republic of China No. 1286718 invention patent "integration of knowledge management system and digital learning system knowledge «integration system and method" reveals a dynamic recommendation module 'when the knowledge management platform is used, it will be based on knowledge and poor materials Correlation and user records recommend the 201131501 learning materials in the learning database; when the digital learning platform is enabled, the relevant knowledge data in the knowledge database will be recommended based on the user's records. However, the main appeal of this case technology lies in the integration of knowledge management and digital learning. It can only passively provide users with knowledge and learning courses, and cannot actively assist users in solving problems. 3. The Republic of China Invention Disclosure No. 2〇〇3〇〇239, “Construction and Analysis Method of Electronic Knowledge Management”, by allowing employees to fill out the Smart Capital Proposal 2 'and use the community to use the website and the Internet Ways to exchange ideas and ideas. However, in this case, the information interpretation of the accumulation and reuse of knowledge is based on the _ word method. The results found by this method are easy to form data noise, giving the user unnecessary answers. Therefore, it is necessary to propose a management information plan that can actively and effectively and accurately assist in solving problems in response to the organizational structure of the enterprise, so as to shorten the time for users to seek answers, and to maximize the benefits of the accumulated knowledge of the organization. SUMMARY OF THE INVENTION Therefore, it is an object of the present invention to provide an active knowledge management system capable of actively and accurately assisting problem solving. Thus, the present invention provides an active knowledge system for problem solving, including a knowledge file database, an automated reply module, and an automated dispatch module. The knowledge file database stores a plurality of knowledge files and feature vectors of the knowledge files. Each knowledge file is numbered by a multi-level work technology mother, and the content of the knowledge file includes an expert who processes the case. Preferably, 201131501, the system further includes - knowledge file management linked to the knowledge file database, the group 'based on the default vocabulary to retrieve keywords from the knowledge files, and calculate the weight of each key skill, and finally The keyword with higher weight is used as the system keyword' and the feature vector of the knowledge file is established according to the system keywords. The knowledge file management module calculates the weight of each __ keyword to refer to the following parameters: the length of the keyword, the length of the longest key bow in the knowledge file of the keyword, and the key of the key gS] appear in the knowledge file The number of occurrences, the number of occurrences of the most frequently occurring keywords in the knowledge file of the key, the total number of knowledge files in the knowledge file database, and the number of knowledge files containing the keyword in the knowledge file database. The automated reply module is coupled to the knowledge file database for receiving a question request command. The content of the command includes the content of the question. Preferably, the question is asked? The internal combination also contains at least one working technology code involved in the problem. "Hai automation reply module converts the problem content into a problem feature vector, and calculates the _' between the problem feature vector and the knowledge m quotation vector in the knowledge file database to increase the inner product value by a preset value The knowledge _ file is provided to the user. Preferably, when the automatic reply module does not have a knowledge file whose inner product value is higher than the preset value, the knowledge class of the upper layer is found according to the working technical code corresponding to the problem. All the knowledge files covered, recalculating the inner product. The automatic dispatch module is linked with the automated reply module, and when the automated reply module calculates the inner product and judges that there is no knowledge file whose inner product value is higher than the preset value, the call is made. The automatic dispatching module, the automated dispatching module searches the knowledge file database for the 201131501 专家 expert who processes the case corresponding to the working technical code according to the working technical code corresponding to the problem, and calculates the degree of conformity of each expert' After the experts sorted according to the degree of conformity, it is recommended to 〇~5 experts and notify the expert to reply. Preferably, the system also includes An expert map, and an expert file is stored in the knowledge file database, and each expert's dimension data is recorded in the expert building; and the automated dispatching module also retrieves the dimension data of the searched expert. Calculate the degree of conformity with the problem based on the dimension data. The dimension data includes the quantitative annual dimension, the professional intensive dimension, and the knowledge added dimension. The formula for calculating the degree of conformity is ·· To the degree of .x professional intensive degree, such as knowledge plus degree, w, u w3 is the pre-determined weight representing the degree of each dimension. Preferably, when the automatic dispatch module is based on the working technology (four) To the expert's, based on the work technology code, find all the knowledge files covered by the knowledge category of the upper layer, and re-search the experts. Preferably, the system also includes - the knowledge map linked to the knowledge file database will be utilized The knowledge file encoded by the working technology code is displayed in a two-dimensional manner, and the first-order encoding corresponds to the second-order encoding, the second-order encoding corresponds to the third-order encoding, and the first-order encoding corresponds to the third. Another aspect of the present invention is to provide a computer program product which, when the computer reads the computer program and executes it, functions as an active knowledge management system for problem solving as described above. Actively includes the following steps: (a) A further object of the present invention is to provide a knowledge management method in which a server executes a 201131501 receiving-question request instruction, the content of the instruction includes a problem content, and converts the content of the question into a question. The feature vector; (b) calculating the inner product between the feature vector of the problem and the feature vector of the majority of the knowledge file in the knowledge file database, and each knowledge file in the knowledge file database is operated by a multi-level function.

The code is numbered, and the content of the knowledge file includes an expert who processes the case; and (C) judges whether there is a knowledge file whose inner product value is higher than a preset value according to the inner product value, and if yes, proceeds to step (d) Provided to the user, if there is no ', then step (e) according to the work technical code corresponding to the problem, search the knowledge file database for the expert who has processed the case corresponding to the technical code, and notify the expert Reply. The utility model has the advantages that the automatic reply module and the automatic dispatch module are combined to design, and the user input questions are automatically queried for accurate answers, or the experts are accurately identified and the experts are notified to reply, thereby actively assisting the problem solving. purpose. The above and other technical contents, features and effects of the present invention will be apparent from the following detailed description of the preferred embodiments of the drawings. Referring to FIG. 1, a preferred embodiment of the active knowledge management system 100 for problem solving of the present invention is suitable for use by a company or various organizations (hereinafter collectively referred to as enterprises), and is constructed on a web server for users to access a computer. 7 Connect the problem to be solved and get an appropriate solution. The system 1 includes a knowledge file management module 1, a knowledge file database 2, a knowledge map 3, an expert map 4, an automated reply module 5, and an automated dispatch module 201131501 group 6. The server has a central processing unit (processing unit material and executing - program H in the memory when the central female Du; Putian secret 4 1 9 々 ' role as the knowledge file S model, and 1, knowledge map 3 , expert group 5, and automatic dispatch module 6. Automated replying model knowledge file database 2, for example, CD-ROM #外_ is the memory of the word server, which is stored in advance, or as a sub-farming 2 The number of eves of the 苇 knowledge case / internal knowledge case (4) is called the knowledge literary cow, the inner valley includes the case data, the expert who handles the case (4) the case is equivalent to being marked in the case. And these:: : 疋 疋 一 一 一 特定 特定 特定 特定 特定 特定 特定 特定 特定 一 一 一 一 一 一 一 一 一 一 一 一 一 一 一 一 一 一 一 一 一 一 一 一 一 一 一 一 一 一 一 一 一 一 一 一 一Level 3 yards, a total of 6 yards can represent the actual 'fourth' content can be designed according to each enterprise. With reference to Figure 2, knowledge file (4) Yu i need to perform pre-work to facilitate the use of subsequent automated reply module 5. Knowledge file management module Group [pre-work includes inclusion - The system_word# 料库u, and the eigenvectors for all the knowledge files in the knowledge file resource library 2 are as follows: Step su-the knowledge file management module 1 according to the preset vocabulary from the knowledge file database 2 The above-mentioned default vocabulary can be constructed according to the existing Chinese vocabulary and the professional domain keywords provided by experts in Keru Enterprise. ^S12 uses [Formula] to calculate each excavation Keyword weights and stored 'highly weighted keywords' are considered more meaningful words for technical solutions. 201131501

[Formula 1] where IMFi, j represents the weight of the keyword in the knowledge file i, represents the length of the keyword j, Li, max represents the length of the longest keyword in the knowledge file i, tfi, j represents the keyword j The number of occurrences in the knowledge document, the taste mu represents the number of occurrences of the most frequently occurring keyword in the knowledge file i 'represents the number of knowledge files containing the keyword j, and N represents the total number of knowledge files in the knowledge file database 2. As the content in the knowledge document database 2 changes, this step can be re-executed periodically, thereby updating the weight value. Step S13 - After the keywords j are sorted according to the weights, the higher weights are taken as system keywords and stored in the system keyword database U. When the system keyword is determined using the above steps, the feature vector of each file can be established.

Step S14 - According to the keyword j in the system keyword A . . u . These keywords are, for example, whose weight in the knowledge file i is wil, Wi2, ... Wij..., and Wjn.

1 double weight, such as [Formula 2] 〇 %)} r ......... [Formula 2]

歹骅 - sigh a question request instruction. The. The content of the instruction includes the input of the vector input 201131501 in the operation screen as shown in Fig. 4 - the problem content (including the subject matter, description) m group of work code involved in the problem. Step S22 - Convert the subject matter and/or the description of the problem into a problem feature vector. The technique for converting text content into feature vectors can be referred to Figure S14 & S15. Which v step S23 - on the other hand, all knowledge files under the knowledge category are found in the knowledge file database based on the work code. Step S24 - Calculate the inner product of the problem feature vector obtained in step S22 and the feature vector of the knowledge file obtained in step S23, and use the inner value as the basis for judging the degree of similarity between the problem and the knowledge file. For example, suppose the system keywords include A, B, c, and D. A question Q proposed by the messenger includes two system keywords A and B, and the weight values calculated by using the formula are 〇4 and 〇4; Then, in step s22, the problem feature vector is obtained as follows: Q= {(A, 0.4), (B, 0.4), (C, 0), (〇, 0)} Assume that the knowledge file found in step S23 has two strokes. The eigenvectors are: D1 = {(A,0.5), (B,0.3),(C,0),(D,0)} 〇2= {(A,0.2),(B,0), (C, 0.3), (D, 0.4)} Then, in step S24, the inner product value of the problem q and the knowledge file m is calculated as 〇·4χ〇.5+().4χ().3+_=() 32, the problem The inner product value of Q and the knowledge component (1) is 0.4χ〇.2 + 0+〇+〇=〇.〇8. Step S25—According to the result obtained in step S24, determining whether there is an inner product value higher than the preset value? If yes, go to step S26, if no, 10 201131501 will enter the flow shown in FIG. 5. The definition of the critical value is to calculate the similarity between the two knowledge files in the knowledge file database 2 in advance, and to make the similarity between 〇 and 1 through normalization; then the similarity obtained from the large After the small sorting, the similarity value with the similarity of about 2% is selected as the critical value when the system screens similar cases. Defining thresholds helps provide relevant cases that users really need to help users solve problems quickly. Step S26—Providing the knowledge file whose inner product value is higher than the preset value to the user reference of φ. The operation screen presented by the system 100 corresponding to this step is as shown in FIG. 6. The user can view the detailed knowledge content and enter the knowledge map 3 by clicking one of the knowledge files. It should be particularly noted here that the knowledge map 3 of the present invention is an effective knowledge query and knowledge presentation tool. When the number of knowledge files is increasing, the way of capturing and presenting knowledge files becomes more important. The knowledge files encoded by the above-mentioned work code coding system are displayed in two dimensions, and the first code is generated. Corresponding to the map of the second three yards (as shown in Figure 7), the map of the second and third yards corresponding to the fourth and fifth yards (as shown in Figure 8), and the first code corresponding to the map of the fourth and fifth yards (as shown in the figure) 9)). These three maps are the basic structure of the knowledge map 3, which allows the reader to quickly understand the knowledge and understand its meaning. In step S27, the automated reply module 5 can also receive a feedback command, which is a user clicking the button "The above suggestions are enough to help me solve the problem" and the "I need expert assistance" button in the figure shown in FIG. Among them - produced. When the user clicks the former button 'automatic reply module 5' to determine the forward feedback command according to the command received by 201131501, the process can be ended; when the user clicks the latter button, the automated reply module 5 will automatically dispatch the call module 6 Then, the steps shown in FIG. 1B are performed by the automated dispatch module 6. In the foregoing step S25_, if the product value of the problem feature vector and the feature vector of each knowledge file is not higher than the preset value, the automated reply module 5 executes the flow shown in FIG. 5, and includes the following steps. Step S31 - The automated reply module 5 finds the knowledge file covered by the upper-layer knowledge category according to the work code ' in the question request instruction, that is, expands the search. Step S32 - calculating the inner product of the problem feature vector and the feature vector of the knowledge file obtained in step S31. (4) S33- According to the calculation result of step S32, it is judged whether there is any inner value higher than the preset value? If yes, proceed to step (10) to provide a knowledge file with an inner product value higher than the preset value to the user for reference. The user can feedback the result of the step S34. Similarly, if the automatic reply module 5 is in step S36. If the forward feedback result is not received, the automatic dispatching module 6 is called; if there is no knowledge file whose inner product value is higher than the preset value, the process proceeds to step S35 'informing the user that there is no similar case for reference, and calling the automatic dispatching mode Group 6. Of course, in the case where there is no knowledge file whose inner product value is higher than the preset value, the present invention can also be designed to perform the steps similar to step 31 again, expand the search range until a knowledge file available for the user is found, or until the determination is made. There is no knowledge file available for user reference in Knowledge Document Database 2 to stop. In the foregoing step S35, and in the case where the step S27 and the step S36 fail to receive the positive 201131501 to the feedback command, the automated dispatch module M6 is called to execute the flow shown in Fig. The process consists of the following steps:

Step S41—the automatic dispatch module 6 searches for the problem request instruction transmitted by the automated reply module $, and searches for the candidate expert corresponding to at least the work code through the expert map 4 according to the work skill code in the instruction. In detail, the knowledge file database 2 stores an expert file. The expert file records the case code and the dimension data (described below) of each case processed by the expert. The expert map 4 of the present invention is an effective expert inquiry and expert resource presentation tool for the user to understand the characteristics of each expert in solving professional problems' and presented in the form of a radar chart or a list as shown in FIG. In this embodiment, an expert who corresponds to the technical code is found by the i-code of the problem, and the expert who has dealt with the relevant case. Step S42 - Confirm whether a candidate expert is found? If yes, the process proceeds to step S43; if the candidate expert cannot be found, the expert corresponding to the work code of the upper layer is found according to the 1 skill code (step S47), that is, the search range is expanded, and the process returns to the step. S42, until it is confirmed that there is no expert related to the problem in the data of the candidate expert or the direct map 4, the second step S43_t finds the candidate expert's reading the foundness information of the expert in this step. In this case, the pre-built expert map 4 is based on the employee's file data, knowledge files (including the case handling personnel) and the knowledge map 3, and is designed for each expert with "years" and " The three attributes of “professional intensive” and “knowledge added value” are quantified as parameters, and the three parameters are called dimension (4), in order to calculate the degree of conformity between experts and problems. In particular, the classification of knowledge maps is the basis for the establishment of expert maps. The expert map structure is based on knowledge maps including professional intensity and knowledge. The value-added dimension is classified by work code. The two kinds of divergence data of the former pumping J are explained as follows. (1) The seniority degree includes designing the seniority, supervising the seniority, and project management, but not limited to this. The annual information is provided by the company. Eight (2) Professional (4) Dimensions can be based on the work skill code corresponding to the type of work handled by the member 1 and the code of the health code. [The hours are used as analytical data. In this embodiment, the professional intensive numerical calculation method is based on the employee's work hours in a certain field divided by the seniority income, and represents the employee's score in this field. Φ (3) Knowledge plus value is the quantification of the value that will occur in knowledge management activities. The so-called "appreciation" means that the output quantity is subtracted from the input quantity, and the amount of knowledge added in the knowledge management process is obtained by using fuzzy words (such as clever terms) through two stages of fuzzy arithmetic calculation. The present invention divides the "original knowledge process" of the problem solving activity into five descriptions. Division is not important, slightly important, important, important and very important; in the "knowledge value added process" description of the word classification, the use of focus group interviews to obtain expert advice 'that is, how many responses to a question Contribution to evaluate _ problem solving activity knowledge value rating. The results of the interviews are divided into five descriptors for the knowledge-added process of problem-solving activities: no contribution, little contribution, contribution, considerable contribution, and very contribution. The measurement of the original knowledge process and the knowledge-added process procedure of the present invention is based on expert judgments of the participants and the administrator, and the expert judgment results are expressed in fuzzy words, so the fuzzy membership function of each word must be established. 14 201131501 The present invention simulates a fuzzy membership function of a fuzzy word by a "bell-shaped function", and the bell-shaped function has a trainable characteristic, and the expression is as shown in [Formula 3]: MF (x) = e 1 σ J [Equation 3] x. : is a variable value in the universe of discourse: membership function value μ: is the middle value of the bell function (mean) σ: is the discriminant value of the bell function (spread) φ by [Equation 3 It can be seen that before the establishment of the fuzzy membership function, the median and spread of each descriptor are known. The present invention describes and presents the knowledge case to the recipient in the form of a network system questionnaire. Interviewers (including questioners and administrators, etc.), and objectively require respondents to make the most appropriate assessment of the original knowledge process and knowledge-added process procedures of knowledge documents and select the most appropriate "original knowledge descriptors" and " Knowledge added descriptive words "After obtaining the results of the interviewee's assessment, the Kohonen Feature Map is used to calculate the median value. The Kohonen Φ Learning Rule consists of two phases, as shown in Equation 4 and Equation 5. 1. The first stage: similarity matching stage (similarity matching) \x-^\ = Min {|^-w;|} [Equation 4] k: kth repetition λ. The normalized value of the cluster w makes the input value The difference between the X and the nearest i-th cluster (<) is minimized. 2. The second stage: the update stage (2011) 15 201131501 + ;; * (Λ: - li;f )

[Formula 5] Y: Appropriate learning coefficient of the kth repetition. In the update phase, the median value of the first cluster, the *th repetition will be applied after the new training data. Clusters that do not have new values will remain unchanged. The Kohonen learning method described above is a cluster (clusteHng) method that determines the intermediate value of the membership function associated with the input and output nodes. The cluster approach is to gather similar data points and determine the median of these data. Each of these similar groups is represented by a vague word, and the inter-value of these groups is the median of the vague words. In order to establish the fuzzy membership function of the bell-shaped function, another parameter-Spread value must be determined. The present invention should find the partial first-nearest_neighb〇rh〇〇d heuristic The different value is as shown in [Formula 6]. σ 丨y [Equation 6] σ: the divergence value w' 'the center value of the i-th cluster ^ nearest · The closest cluster center value r: the overlap ratio of two adjacent fuzzy words when the knowledge bonus mode is in front Already established, the amount of knowledge change between input and output will be calculated next. The invention presents and calculates the change range by using the two stages of the knowledge value-adding process, the original knowledge process and the knowledge value-adding process, that is, using fuzzy "fuzzy anthmetic" to blur the multiplication process to the fuzzy set process. Calculation. Assume that the first knowledge management activity of 16 201131501 is taken as an example. The fuzzy words of the original knowledge process and the knowledge bonus process are respectively 3 and the heart is added to the knowledge calculation formula as shown in [Formula 7]: [Equation 7] Estimation of knowledge plus In the value process, the knowledge value of the knowledge management activities increases by 3: the fuzzy words of the original knowledge process: the fuzzy words of the knowledge value-added process in order to calculate the knowledge value of multiple knowledge management activities, the fuzzy words must be blurred. After analyzing the existing defuzzification method, the calculation efficiency of Gaussian fuzzy membership zero function is considered, and the "maximum intermediate method (max_〇f嶋(4)" is used as the method of defuzzification. The addition of knowledge addition is calculated as shown in [8] κνΑ〇,αΙ = ΣκνΑ [Formula 8] Ruler: «The knowledge plus value of the knowledge management activities is the sum of the factory: the knowledge of the knowledge management activities, the value of the knowledge added to the above knowledge management activities Is the total amount of value used to present the knowledge community.

In general, the knowledge value-added dimension is used to judge whether the experts are willing to share the value-added judgment. If an expert is very happy to share, the expert's score will be improved in the calculation of the knowledge added value. For example, as a whole, it is assumed that the automatic dispatching module 6 wants to find an expert who can solve the user's question from the technical code 110Α00, and a total of j 3 candidates in the work skill code _ select the expert' this step to read the base top direction斛 斛 斛 设计 设计 设计 设计 设计 设计 设计 设计 设计 设计 设计 设计 设计 设计 设计 2011 2011 2011 2011 2011 2011 2011 2011 2011 2011 2011 2011 2011 2011 2011 2011 2011 2011 2011 2011 2011 2011 2011 2011 2011 2011 2011 2011 2011 2011 2011 2011 2011 2011 2011 2011 2011 2011 2011 2011 2011 2011 The degree of conformity between the expert and the user's question, after sorting the experts according to the highest to the lowest, it is recommended to 〇~5 digits (step S45) 'Also send an email to the expert 8 (step). The user can click on the system! 〇〇 The recommended experts enter the expert map 4 as shown in Figure u to view the expert information. Following the above example, the degree of conformity of each candidate expert can be calculated by the following [3] and [Formula 4]: Compliance degree + "2X professional intensive degree + . Knowledge plus degree [3] =W4 X Design Years + w5 χ Supervising $胄+ γ χ stomach

...L type 4J

Wi, W2, and W3 in Equation 3 represent the weights of the respective degrees of orientation, and each weight is defined in advance by the administrator and stored and set. The degree of capital in [Formula 3] is calculated by [Formula 4], and W4, Ws, and w0 are also pre-existing and stored by the administrator. Through the above formula, with Wi = 〇2, W2=〇4 and W3~0·4 and W4=〇.33, W5=0_34 and W6=〇.33 as examples, the scores of the candidate experts are: Candidate 1 candidate Expert 2 : 0.2 X (0.33 X10 + 0.34 x 0 + 0.33 x 3) + 0.4 χ 56.8 + 0.4 χ 0 == 23.578 0.2 χ (0.33 χ 20 + 0.34 χ 10 + 0.33 χ 0) + 0.4 χ 100 + 0.4 χ 〇42 18 201131501 Candidate Expert 3 : 0·2χ(0.33χ0 + 0·34χ5 + 0.33χ0) + 0.4χ72.6 + 0·4Χ0 = 29.38 Therefore, when the questioner asks, he cannot get a response in the automated reply module 5 or When the replies are not used, in addition to posting the questions and waiting for the enthusiasts to reply, you can also seek expert assistance in the field through the automated dispatch module 6. In this way, not only can the time for the responder to wait for the reply of the person to be saved, but also the assistance of the expert in the field judged by the automated dispatch module 6 can be obtained, so that the reply can be more accurate. In summary, the present invention is an active knowledge management system for problem solving. The method and method can use the multi-level technical code to pre-code the case according to the organizational structure of the enterprise, and utilize the automated reply module 5 and the automation. The dispatching module 6 solves the problem raised by the user actively, effectively and accurately, so as to shorten the time for the user to seek the answer, and let the accumulated knowledge of the enterprise organization maximize the benefits and fully inherit the original, so the invention can be achieved. The purpose. However, the above is only the preferred embodiment of the present invention, and the scope of the present invention is not limited thereto, that is, the simple equivalent changes and modifications made by the present invention in accordance with the invention and the description of the invention. , are still within the scope of this patent. BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram showing a preferred embodiment of an active knowledge management system for solving a problem in the present invention; FIG. 2 is a diagram illustrating the operation of a knowledge file management module in the embodiment. FIG. 3 is a partial flow chart showing the flow of execution of the automatic reply module in the embodiment; 19 201131501 FIG. 4 is a schematic diagram of an operation screen of the user-proposed problem; FIG. Another part of the flow of the process; a schematic diagram of the operation screen for providing the query result by the automated reply module; FIG. 7 to FIG. 9 are schematic diagrams of various knowledge maps in the preferred embodiment, FIG. 10 is a schematic view of the preferred embodiment. A flow chart of the flow executed by the automated dispatch module; and FIG. 11 is a schematic diagram of the query of the expert map in the preferred embodiment.

20 201131501 [Description of main component symbols] 100... •...Proactive problem solving 6... .......Automatic dispatching modular knowledge management system 7 ··· ....... (user) computer 1 ....... •...Knowledge File Management Module 8··· ......(Expert) Computer 11 ······...System Keyword Database S11~S 15Step 2... •...Knowledge Document Library S21 ~ S27 Step 3 ....... — Knowledge Map S31 ~ S36 Step 4 ....... Special Map S41 ~ S47 Step 5 ....... •...Automatic Reply Module

twenty one

Claims (1)

201131501 VII. Patent application scope: 1. An active knowledge management system for problem solving, comprising: a knowledge file database 'storing a plurality of knowledge files and feature vectors of the knowledge files, each knowledge file having more than one The working technology code of the class is numbered, and the content of the knowledge file includes an expert who processes the case; an automated reply module is connected with the knowledge file database for receiving a question request instruction, the content of the instruction includes the content of the question, The automated reply module converts the problem content into a problem feature vector, and calculates an inner product between the problem feature vector and the feature vector of the knowledge file in the knowledge file database, and the inner product value is higher than a preset value. The knowledge file provides φ to the user; and - the automated dispatch module is coupled to the automated reply module, and when the automated back f module calculates (10) and determines that the _ value is higher than the knowledge file of the preset 5, the call is made. The automatic dispatching module, the automated dispatching module is based on the problem The technical code is in the knowledge file database I:::::: the expert of the case, and informs the proactive knowledge system for problem solving, as described in item 1, wherein the problem request instruction The content also includes at least one working technical code involved in the question, and the working technical code of the automation enters (4). In the instruction, please refer to the knowledge management system described in item i of the full-time project, including the expert map, and the active knowledge holder has the expert #索^ The broadcast case 0 records the divergence data of each expert; 22 201131501 The automated dispatch module also retrieves the dimension data of the searched experts and calculates the degree of conformity with the problem based on the β-Hair dimension data. 4. The active knowledge management system for problem solving according to the scope of patent application 帛3, wherein the automatic dispatch module sorts the searched experts according to the obtained degree of conformity, and at most informs the sorting. The top five experts responded. 5. According to the active knowledge management system for problem solving described in item 3 of the scope of the patent application, wherein the dimension data includes a quantitative annual dimension and a professional intensive degree related to the number of hours worked, And knowledge added value. 6. According to the patent, the active knowledge for problem solving, as described in item 5 of the patent scope, is the rationale system, where 'the automatic dispatch module uses the following formula to calculate the expert complexity D degree. Compliance = % 2 degrees of professional intensive knowledge plus degree; ...%, 霄2 and ^ are pre-determined weights representing the respective degrees of orientation. 7. According to the active knowledge management system for problem solving described in the scope of patent application 帛1, where the automatic reply module determines that there is no knowledge file with the inner product value high 2 the preset value, according to the problem The corresponding work technology = find all the knowledge files covered by the knowledge category of the previous level, and recalculate the inner product. 8. = According to the patent application, the active knowledge for problem solving is: the rational system, in which, when the automatic dispatch module searches for two experts according to the working technology code, it is based on the technology. The code finds all the knowledge files of the upper-level knowledge category/letter and re-searches the experts. 9. According to the scope of the patent application scope, the active knowledge for problem solving 23 201131501 management system also includes a knowledge (four) map connected with the knowledge file database, which will be coded by using the working technical code. The knowledge of the . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ~ 1〇·依射, please refer to the active knowledge management system for problem solving described in item 1 of the patent scope, and also include a knowledge management module linked to the knowledge file database, based on a preset The thesaurus draws the key d from the knowledge-+^ as knowledge files, and calculates the weight of each keyword. The most M ^ ^ weights are compared to the keywords as system keywords, and are read according to the market. && The feature vector of these knowledge files is established according to the keywords of the system. 1111. According to the scope of claim 9th, the $ π 4 knowledge management system for problem solving, wherein the knowledge file describes the word « _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ The length of the keyword, the length of the longest keyword in the knowledge file in which the keyword is located, the number of occurrences of the keyword in the knowledge file in which it is located, the knowledge of the keyword ^ ^ a _ the occurrence of the most frequently occurring keyword - the number of people, the total number of knowledge files in _ ^ ^ in the knowledge file database, and the number of knowledge files containing the keyword in the knowledge file database. 12. A computer program product, when a computer program in the computer program product is sold and executed, the function is as described in any one of the application scopes 1 to 1 Proactive knowledge of solution] 3 · An active y_, method for problem solving, executed by a server, and includes the following steps: (a) receiving a request for a request, the content of the instruction includes the problem Combine ' and convert the content of the question into a question vector of eigen; 24 201131501 知 』 b) 4 eigenvalues of the phase problem and the majority of the knowledge file database: the inner product between the feature vectors of the file, the knowledge file data The parent-knowledge organization w 々1 file u—multi-level work technical code is numbered, and the content of the knowledge file includes an expert who processes the case; and) the internal product value is used to determine whether the inner product value is higher than a preset The value of the document 'if it exists' is carried out in step (4) to the user, if ^ ' 1 彳 仃 (4) (e) according to the work technology code corresponding to the problem -, ^ (4) work technology mother corresponding The expert of the case, inform the expert to reply. 14:: According to: the scope of patents...proactive method for problem solving: management method', wherein the step (° also includes the dimension data of the searched by:!) based on the dimension data Calculate the relationship between the problem and the problem. The degree of the data recorded in the knowledge database is based on the knowledge management method described in Item 13 of the scope of the patent application, and includes a knowledge management method. Pre-processing p / outside, according to a pre-% Gusi = from the material knowledge file to fetch keywords, and calculate the weight of each key word, dare to extract the higher weight of the keyword as a system keyword, 5 Hai And other system keywords to establish the characteristics of these knowledge files to 3^ 25