Search sorting method and device

Abstract

The present application provides a search ranking method and device that can be used in the financial field or other fields. The method includes: receiving a target search statement; obtaining multiple recall statements from a preset operation and maintenance knowledge base based on the target search statement; determining the operation and maintenance knowledge words corresponding to the target search statement and each recall statement, as well as the word weight and word vector of each operation and maintenance knowledge word based on a pre-stored operation and maintenance knowledge graph, a word vector set, the target search statement, and each recall statement; and determining and outputting the ranking results of the multiple recall statements based on the word weight and word vector of each operation and maintenance knowledge word. The present application can improve the efficiency and accuracy of search ranking, thereby improving the reliability of data center operation and maintenance.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a search and sorting method and device. Background Art

[0002] With the development of society and the economy, the construction of computing infrastructure, such as data centers, has become a top priority. Operations and maintenance, as a crucial component of data centers, provides crucial support for their smooth operation. Operations and maintenance are a technology- and information-intensive field, characterized by a broad scope of knowledge, diverse areas, complex problems, and difficulties in entry, in-depth understanding, and the dissemination of knowledge. Currently, addressing the challenges of knowledge accumulation, search, and dissemination in the field of operations and maintenance presents significant challenges.

[0003] Daily operations and maintenance generate a vast amount of valuable system documentation, plans, problem-solving solutions, regulations, and operating manuals. Most of this documentation is scattered among various experts or stored in tools like Confluence. Acquiring knowledge through expert communication is inefficient, while searching and sorting knowledge through Confluence-like tools suffers from low accuracy. Summary of the Invention

[0004] In response to at least one problem in the prior art, the present application proposes a search ranking method and device, which can improve the efficiency and accuracy of search ranking, and thereby improve the reliability of data center operation and maintenance.

[0005] In order to solve the above technical problems, this application provides the following technical solutions:

[0006] In a first aspect, the present application provides a search ranking method, comprising:

[0007] receiving a target search statement;

[0008] Obtaining multiple recall statements from a preset operation and maintenance knowledge base according to the target search statement;

[0009] Determine the operation and maintenance knowledge words corresponding to the target search statement and each recall statement, as well as the word weight and word vector of each operation and maintenance knowledge word, based on the pre-stored operation and maintenance knowledge graph, the word vector set, the target search statement, and each recall statement;

[0010] According to the word weight and word vector of each operation and maintenance knowledge word, the sorting results of the multiple recall statements are determined and output.

[0011] Furthermore, the operation and maintenance knowledge words corresponding to the target search statement and each recall statement, as well as the word weight and word vector of each operation and maintenance knowledge word, are determined based on the pre-stored operation and maintenance knowledge graph, the word vector set, the target search statement, and each recall statement, including:

[0012] Determining the operation and maintenance knowledge words corresponding to the target search statement and each recall statement, as well as the word weight of each operation and maintenance knowledge word, based on the pre-stored operation and maintenance knowledge graph, the target search statement, and each recall statement;

[0013] Determine the word vector for each operation and maintenance knowledge word based on the pre-stored word vector set.

[0014] Furthermore, the determining and outputting of the sorting results of the plurality of recall statements based on the word weight and word vector of each operation and maintenance knowledge word includes:

[0015] Calculate similarity based on the word weights and word vectors of the operation and maintenance knowledge words in the target search sentence and the word weights and word vectors of the operation and maintenance knowledge words in each recall sentence;

[0016] According to the similarity calculation result of each recall sentence, the ranking result of the plurality of recall sentences is determined and output.

[0017] Furthermore, before obtaining a plurality of recall statements from a preset operation and maintenance knowledge base according to the target search statement, the method further includes:

[0018] Obtain original operation and maintenance knowledge documents;

[0019] Storing the plurality of operation and maintenance knowledge statements in the original operation and maintenance knowledge document in the preset operation and maintenance knowledge base;

[0020] The recall statement is an operation and maintenance knowledge statement corresponding to the target search statement obtained from the preset operation and maintenance knowledge base according to a recall algorithm.

[0021] Furthermore, obtaining multiple recall statements from a preset operation and maintenance knowledge base based on the target search statement includes: obtaining multiple recall statements from a preset operation and maintenance knowledge base based on a multi-way recall algorithm and the target search statement, and the multi-way recall algorithm includes: TFIDF and BM25 algorithms.

[0022] Furthermore, before obtaining the word weight and word vector of each operation and maintenance knowledge word from the pre-stored operation and maintenance knowledge graph and word vector set, the method further includes:

[0023] Obtaining an original operation and maintenance knowledge document, where the original operation and maintenance knowledge document includes multiple operation and maintenance knowledge statements;

[0024] Performing word segmentation processing on the original operation and maintenance knowledge document to obtain a plurality of original operation and maintenance knowledge words;

[0025] Applying the TextRank algorithm and the plurality of original operation and maintenance knowledge words to determine the word weight of each original operation and maintenance knowledge word;

[0026] The operation and maintenance knowledge graph is obtained based on each original operation and maintenance knowledge word and its respective word weight, wherein the operation and maintenance knowledge graph includes: nodes, node weights and directed edges between nodes, nodes represent original operation and maintenance knowledge words, and directed edges represent the contextual relationship between original operation and maintenance knowledge words in operation and maintenance knowledge sentences.

[0027] Furthermore, after performing word segmentation processing on the original operation and maintenance knowledge document to obtain a plurality of original operation and maintenance knowledge words, the method further includes:

[0028] Perform embedding vectorization processing on each original operation and maintenance knowledge word to obtain the word vector of each original operation and maintenance knowledge word;

[0029] The word vector set is obtained according to each original operation and maintenance knowledge word and its respective word vector.

[0030] In a second aspect, the present application provides a search ranking device, comprising:

[0031] A receiving module, configured to receive a target search statement;

[0032] A recall module, configured to obtain a plurality of recall statements from a preset operation and maintenance knowledge base according to the target search statement;

[0033] A determination module is used to determine the operation and maintenance knowledge words corresponding to the target search statement and each recall statement, as well as the word weight and word vector of each operation and maintenance knowledge word based on the pre-stored operation and maintenance knowledge graph, the word vector set, the target search statement and each recall statement;

[0034] The sorting module is used to determine and output the sorting results of the multiple recall statements based on the word weight and word vector of each operation and maintenance knowledge word.

[0035] Furthermore, the sorting module includes:

[0036] A scoring unit, configured to calculate similarity based on the word weights and word vectors of the operation and maintenance knowledge words of the target search sentence and the word weights and word vectors of the operation and maintenance knowledge words of each recall sentence;

[0037] The determining unit is configured to determine and output a ranking result of the plurality of recall statements based on a similarity calculation result of each recall statement.

[0038] Furthermore, the search ranking device further includes:

[0039] Acquisition module, used to obtain original operation and maintenance knowledge documents;

[0040] A storage module, configured to store the plurality of operation and maintenance knowledge statements in the original operation and maintenance knowledge document in the preset operation and maintenance knowledge base;

[0041] The recall statement is an operation and maintenance knowledge statement corresponding to the target search statement obtained from the preset operation and maintenance knowledge base according to a recall algorithm.

[0042] In a third aspect, the present application provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the search and sorting method when executing the program.

[0043] In a fourth aspect, the present application provides a computer-readable storage medium having computer instructions stored thereon, which implement the search ranking method when the instructions are executed.

[0044] As can be seen from the above technical solution, the present application provides a search sorting method and device. The method includes: receiving a target search statement; obtaining multiple recall statements from a preset operation and maintenance knowledge base according to the target search statement; determining the operation and maintenance knowledge words corresponding to the target search statement and each recall statement, as well as the word weight and word vector of each operation and maintenance knowledge word according to the pre-stored operation and maintenance knowledge graph, word vector set, the target search statement and each recall statement; determining and outputting the sorting results of the multiple recall statements according to the word weight and word vector of each operation and maintenance knowledge word, which can improve the efficiency and accuracy of search sorting, and thus improve the reliability of data center operation and maintenance; specifically, it can quickly share operation and maintenance knowledge and accurately locate problems; it can lower the operation and maintenance threshold, improve operation and maintenance efficiency, save labor costs, and ensure the stable operation of the data center. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0046] Figure 1 Schematic diagram of the search and sorting method in the embodiment of the present application;

[0047] Figure 2 is a flowchart of a search and sorting method in another embodiment of the present application;

[0048] Figure 3 0031 to 0034 of the search and sorting method in an embodiment of the present application;

[0049] Figure 4 This is a schematic diagram of the structure of the search and sorting device in an embodiment of the present application;

[0050] Figure 5 This is a schematic block diagram of the system structure of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0051] In order to enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0052] In order to facilitate the understanding of this solution, first, the technical content related to this solution is explained.

[0053] Embedding: It is a way to convert discrete variables into continuous vector representations; it not only reduces the dimensionality of the discrete space, but also enables the discovery of more similar statements in high-dimensional space.

[0054] TFIDF: A commonly used weighting technique used in information retrieval and text mining, it is used to calculate the importance of a word to a document. The importance of a word increases with its frequency in a document and decreases with its frequency in a document library. TFIDF = TF * IDF, where:

[0055]

[0056]

[0057] BM25: The final query score is obtained by calculating the sum of the scores of the morphemes (Chinese word segmentations) in the query statement in the documents, Score(Q, d). There are many ways to calculate the weights of morphemes and documents, and the IDF algorithm is usually used.

[0058]

[0059]

[0060]

[0061] Among them, Q represents the query statement, d represents the recall statement, and q i represents the i-th word, n represents the number of query words, W i represents the weight between the i-th word recall statement, K represents the adjustment factor, q represents the query statement, k1, b and k2 are adjustment factors; R(qi ,d) represents the correlation score between the search statement and the recall statement, q i represents a query statement, d represents a recall statement; f i Indicates the frequency of morphemes (words) appearing in q, qf i It represents the frequency of morphemes (words) appearing in the recall sentences. It is usually set based on experience that k1=k2=2 and b=0.75; dl is the document length, and avgdl is the average document length in the document library.

[0062] Elasticsearch (ES) is a Lucene-based search server that provides a distributed, multi-user, full-text search engine with a RESTful interface. ES is a distributed storage tool for log storage, storing data in the form of key-value pairs.

[0063] Currently, common ways of sharing and storing knowledge are divided into the following categories:

[0064] 1. Experts store "system documents, plans, solutions, regulations, and operating manuals" based on experience, and then use tools similar to "Everythings" for keyword retrieval during searches. This expert storage method has the problems of fragmented storage and difficulty in sharing. Searches are based solely on keywords, so the accuracy is low.

[0065] 2. Enterprises use tools like Conflunce to store and share information. Conflunce uses the Lucene algorithm to provide search capabilities. Because the Lucene algorithm is based on an inverted index, shared document tools can only search and sort by keyword scoring, with low accuracy.

[0066] Based on this, the present application provides a search ranking method and device, which can realize the high-efficiency and high-accuracy construction, storage and search of the operation and maintenance knowledge base; through multi-way recall and graph network structure, the weights of the search statements and recall statements are obtained, and the recall statements are finally sorted and the sorting results are returned. While improving the efficiency of knowledge sharing, the retrieval efficiency is improved and the operation and maintenance threshold is lowered. Knowledge construction is to store and vectorize the original operation and maintenance knowledge; rough knowledge sorting is to perform preliminary recall of massive operation and maintenance knowledge; fine knowledge sorting is to calculate the importance of sentence vectors with a graph network algorithm, and further optimize the search ranking results based on the sentence similarity. The operation and maintenance knowledge is itemized and vectorized, and the retrieval accuracy is improved through rough sorting and fine sorting methods; the fine sorting method uses a graph network structure to calculate the importance of word nodes in the sentence, and when comparing sentence similarity, the key words are highlighted; it is convenient for knowledge sharing and storage, reduces the operation and maintenance threshold, and improves the operation and maintenance efficiency; based on the application of the multi-way recall algorithm, this solution can improve the recall accuracy while streamlining the number of recall statements output through fine knowledge sorting.

[0067] To improve the efficiency and accuracy of search ranking, and thereby enhance the reliability of data center operations and maintenance, embodiments of the present application provide a search ranking device, which may be a server or client device. The client device may include a smartphone, a tablet electronic device, a network set-top box, a portable computer, a desktop computer, a personal digital assistant (PDA), an in-vehicle device, and a smart wearable device. The smart wearable device may include smart glasses, smart watches, and smart bracelets.

[0068] In actual applications, the search and sorting portion can be performed on the server side as described above, or all operations can be completed on the client device. The specific selection can be based on the processing capabilities of the client device and the limitations of the user's usage scenario. This application does not impose any restrictions on this. If all operations are completed on the client device, the client device may also include a processor.

[0069] The client device may include a communication module (i.e., a communication unit) that can establish a communication connection with a remote server to implement data transmission with the server. The server may include a server on the task scheduling center side, and in other implementation scenarios, may also include a server on an intermediate platform, such as a server on a third-party server platform that has a communication link with the task scheduling center server. The server may include a single computer device, a server cluster consisting of multiple servers, or a server structure of a distributed device.

[0070] The server and the client device may communicate using any suitable network protocol, including network protocols not yet developed on the date of filing this application. Examples of such network protocols include TCP / IP, UDP / IP, HTTP, and HTTPS. Furthermore, examples of such network protocols include RPC (Remote Procedure Call Protocol) and REST (Representational State Transfer) protocols, which are used on top of the aforementioned protocols.

[0071] It should be noted that the search ranking method and device disclosed in this application can be used in the field of financial technology, and can also be used in any field other than the field of financial technology. The application field of the search ranking method and device disclosed in this application is not limited.

[0072] The details are described in detail through the following embodiments.

[0073] In order to improve the efficiency and accuracy of search sorting and thus improve the reliability of data center operation and maintenance, this embodiment provides a search sorting method in which the execution subject is a search sorting device, which includes but is not limited to a server, such as Figure 1 As shown, the method specifically includes the following contents:

[0074] Step 100: Receive a target search statement.

[0075] Specifically, the search ranking device can receive the target search statement sent by the front end.

[0076] Step 200: Obtain multiple recall statements from a preset operation and maintenance knowledge base according to the target search statement.

[0077] Specifically, according to the recall algorithm, the operation and maintenance knowledge statement corresponding to the target search statement can be obtained from the preset operation and maintenance knowledge base as the recall statement; the preset operation and maintenance knowledge base can pre-store a large number of operation and maintenance knowledge statements; step 200 can be equivalent to the process of rough knowledge sorting.

[0078] Step 300: Determine the operation and maintenance knowledge words corresponding to the target search statement and each recall statement, as well as the word weight and word vector of each operation and maintenance knowledge word, based on the pre-stored operation and maintenance knowledge graph, the word vector set, the target search statement and each recall statement.

[0079] Specifically, the pre-stored operation and maintenance knowledge graph may include: the correspondence between operation and maintenance knowledge words and word weights; the pre-stored word vector set may include: the correspondence between operation and maintenance knowledge words and word vectors.

[0080] Step 400: Determine and output the ranking results of the multiple recall statements based on the word weight and word vector of each operation and maintenance knowledge word.

[0081] Specifically, the multiple recall statements can be output and displayed according to the sorting results; in order to reduce redundant data and improve the efficiency of data transmission, further, after determining the sorting results of the multiple recall statements, the recall statement data of the top N can be displayed. N can be set according to actual conditions, and this application does not impose any restrictions on this. Steps 200 to 300 can be equivalent to the process of knowledge sorting. The amount of data of the recall statements is very large (up to hundreds or even thousands), so further screening is required; the word weights of each operation and maintenance knowledge word in the search statement and the recall statement are searched in the graph structure, i.e., the operation and maintenance knowledge graph (the default weight that cannot be recalled is the inverse of the number of nodes in the graph), and the similarity between the word vector and word weight of the search statement and the word vector and word weight of the recall statement is calculated, and the final score is used for sorting.

[0082] In order to improve the accuracy of obtaining word vectors and word weights, in one embodiment of the present application, step 300 includes:

[0083] Step 301: According to the pre-stored operation and maintenance knowledge graph, the target search statement and each recall statement, determine the operation and maintenance knowledge words corresponding to the target search statement and each recall statement and the word weight of each operation and maintenance knowledge word.

[0084] Specifically, the target search statement can be segmented to obtain the operation and maintenance knowledge words corresponding to the target search statement, and it can be determined whether the operation and maintenance knowledge words exist in the pre-stored operation and maintenance knowledge graph. If so, the word weight of the operation and maintenance knowledge words is obtained from the pre-stored operation and maintenance knowledge graph. Otherwise, the word weight of the operation and maintenance knowledge words corresponding to the target search statement can be set to the inverse of the number of nodes in the graph, which neither highlights nor weakens the importance of the words; the original operation and maintenance knowledge words belonging to the recall statement can be obtained from the pre-stored operation and maintenance knowledge graph, and the original operation and maintenance knowledge words are used as the operation and maintenance knowledge words corresponding to the recall statement, and the word weight corresponding to the operation and maintenance knowledge words is obtained from the pre-stored operation and maintenance knowledge graph.

[0085] Step 302: Determine the word vector of each operation and maintenance knowledge word based on the pre-stored word vector set.

[0086] Specifically, a record in which the original operation and maintenance knowledge word is the operation and maintenance knowledge word can be obtained from the pre-stored word vector set, and the word vector in the record is used as the word vector of the operation and maintenance knowledge word.

[0087] Using only one recall algorithm may have loopholes. In order to ensure the recall rate of data and improve the reliability of recall statements, in one embodiment of the present application, step 200 includes: obtaining multiple recall statements from a preset operation and maintenance knowledge base according to a multi-way recall algorithm and the target search statement, and the multi-way recall algorithm includes: TFIDF and BM25 algorithms.

[0088] To improve the reliability of recall sentence ranking results, see Figure 2 In one embodiment of the present application, step 400 includes:

[0089] Step 401: Calculate similarity based on the word weights and word vectors of the operation and maintenance knowledge words in the target search statement and the word weights and word vectors of the operation and maintenance knowledge words in each recall statement.

[0090] Step 402: According to the similarity calculation result of each recall sentence, determine the sorting result of the plurality of recall sentences and output it.

[0091] Specifically, the similarity calculation result of the recall sentence can be used as the ranking score of the recall sentence, and the recall sentence can be sorted from high to low according to the ranking score.

[0092] In order to improve the reliability of storage of operation and maintenance knowledge statements, and thus improve the reliability of subsequent recall of operation and maintenance knowledge statements, in one embodiment of the present application, before step 200, the following steps are further included:

[0093] Step 021: Obtain the original operation and maintenance knowledge document.

[0094] Specifically, the original operation and maintenance knowledge documents may include: system documents, plans, problem solutions, regulations and operation manuals of the data center.

[0095] Step 022: Store multiple operation and maintenance knowledge statements in the original operation and maintenance knowledge document in the preset operation and maintenance knowledge base; the recall statement is an operation and maintenance knowledge statement corresponding to the target search statement obtained from the preset operation and maintenance knowledge base according to the recall algorithm.

[0096] Specifically, the original knowledge document can be split into segments and strips; the knowledge strips can be stored in ES by row, and ES can be replaced by SOLR search application server, MYSQL and ORACLE database, etc.; a section of the original operation and maintenance knowledge document can be used as an operation and maintenance knowledge statement.

[0097] In order to improve the accuracy of obtaining the operation and maintenance knowledge graph, see Figure 3 In one embodiment of the present application, before step 300, the method further includes:

[0098] Step 0031: Obtain an original operation and maintenance knowledge document, which includes multiple operation and maintenance knowledge statements.

[0099] Step 0032: Perform word segmentation processing on the original operation and maintenance knowledge document to obtain multiple original operation and maintenance knowledge words.

[0100] Step 0033: Apply the TextRank algorithm and the multiple original operation and maintenance knowledge words to determine the word weight of each original operation and maintenance knowledge word.

[0101] Specifically, in the original operation and maintenance knowledge document, key words are more prominent. Therefore, extracting the word weights of words based on the original operation and maintenance knowledge document can better highlight the importance of the words; the original operation and maintenance knowledge words are the words obtained after word segmentation processing of the original operation and maintenance knowledge document.

[0102] Step 0034: The operation and maintenance knowledge graph is obtained based on each original operation and maintenance knowledge word and its respective word weight, wherein the operation and maintenance knowledge graph includes: nodes, node weights and directed edges between nodes, nodes represent original operation and maintenance knowledge words, and directed edges represent the contextual relationship between the original operation and maintenance knowledge words in the operation and maintenance knowledge sentences.

[0103] Specifically, the line from the word before the current word to the current word is called the in-degree (in-edge), and the line between the current word and the word after the current word is called the out-degree (out-edge). The operation and maintenance knowledge graph includes: each original operation and maintenance knowledge word and its respective word weight. The original operation and maintenance knowledge document can be decomposed into words and formed into a graph structure. The importance of the node weight is calculated by the number of in-degree and out-degree of each word. The specific description is as follows:

[0104] a) Segment the original operation and maintenance knowledge document and generate the initial graph structure; set the initial weight of each node (the default is the inverse of the number of nodes in the graph structure) to WS (V j ) is set to the inverse of the number of nodes in the graph structure; the out-degree weight w of each node is set jk The default value is the number of out-degrees of the node, that is, the number of connections from the node to other nodes; set the in-degree weight w of each node ij The default value is the number of in-degrees of the node, that is, the number of links from other nodes to the node.

[0105] b) Obtain the word weight of each node according to the calculation formula of word importance; the calculation formula of importance is as follows:

[0106]

[0107] in, represents the sum of all out-degree weights of each node, represents the importance weight of the node itself, d represents the damping coefficient, which is set to 0.85 by default; after several rounds of iteration, the word weight of each node is calculated.

[0108] c) Based on the word weight of each node, the original sentence graph structure, namely the above-mentioned operation and maintenance knowledge graph, is obtained.

[0109] In order to improve the reliability of obtaining the word vector set, in one embodiment of the present application, before step 300, the following steps are further included:

[0110] Step 311: Perform embedding vectorization processing on each original operation and maintenance knowledge word to obtain a word vector for each original operation and maintenance knowledge word.

[0111] Step 312: Obtain the word vector set based on each original operation and maintenance knowledge word and its respective word vector.

[0112] Specifically, the word vector set includes: each original operation and maintenance knowledge word and its respective word vector.

[0113] From the software level, in order to improve the efficiency and accuracy of search sorting, and thus improve the reliability of data center operation and maintenance, this application provides an embodiment of a search sorting device for implementing all or part of the content of the search sorting method, see Figure 4 , the search and ranking device specifically includes the following contents:

[0114] The receiving module 10 is configured to receive a target search statement.

[0115] The recall module 20 is used to obtain multiple recall statements from a preset operation and maintenance knowledge base according to the target search statement.

[0116] The determination module 30 is used to determine the operation and maintenance knowledge words corresponding to the target search statement and each recall statement, as well as the word weight and word vector of each operation and maintenance knowledge word based on the pre-stored operation and maintenance knowledge graph, the word vector set, the target search statement and each recall statement.

[0117] The sorting module is used to determine and output the sorting results of the multiple recall statements based on the word weight and word vector of each operation and maintenance knowledge word.

[0118] In one embodiment of the present application, the sorting module includes:

[0119] The scoring unit is used to calculate similarity based on the word weights and word vectors of the operation and maintenance knowledge words of the target search sentence and the word weights and word vectors of the operation and maintenance knowledge words of each recall sentence.

[0120] The determining unit is configured to determine and output a ranking result of the plurality of recall statements based on a similarity calculation result of each recall statement.

[0121] In one embodiment of the present application, the search ranking device further includes:

[0122] The acquisition module is used to obtain original operation and maintenance knowledge documents.

[0123] The storage module is used to store the plurality of operation and maintenance knowledge statements in the original operation and maintenance knowledge document in the preset operation and maintenance knowledge base.

[0124] The recall statement is an operation and maintenance knowledge statement corresponding to the target search statement obtained from the preset operation and maintenance knowledge base according to a recall algorithm.

[0125] The embodiment of the search ranking device provided in this specification can be specifically used to execute the processing flow of the embodiment of the above-mentioned search ranking method. Its functions will not be described in detail here, and reference can be made to the detailed description of the embodiment of the above-mentioned search ranking method.

[0126] From the above description, it can be seen that the search and sorting method and device provided in this application can improve the efficiency and accuracy of search and sorting, and thus improve the reliability of data center operation and maintenance; specifically, it can not only quickly share operation and maintenance knowledge, but also accurately locate problems; it can lower the operation and maintenance threshold, improve operation and maintenance efficiency, save labor costs, and ensure the stable operation of the data center.

[0127] From a hardware perspective, in order to improve the efficiency and accuracy of search and ranking, and thereby improve the reliability of data center operations and maintenance, the present application provides an embodiment of an electronic device for implementing all or part of the content of the search and ranking method. The electronic device specifically includes the following content:

[0128] A processor, a memory, a communications interface, and a bus; wherein the processor, the memory, and the communications interface communicate with each other via the bus; the communications interface is used to implement information transmission between the search ranking device and related devices such as user terminals; the electronic device can be a desktop computer, a tablet computer, a mobile terminal, etc., but this embodiment is not limited thereto. In this embodiment, the electronic device can be implemented with reference to the embodiments for implementing the search ranking method and the embodiments for implementing the search ranking device, the contents of which are incorporated herein and repeated parts are not repeated.

[0129] Figure 5 Schematic block diagram of the system structure of the electronic device 9600 according to an embodiment of the present application. Figure 5As shown, the electronic device 9600 may include a central processing unit 9100 and a memory 9140; the memory 9140 is coupled to the central processing unit 9100. It is worth noting that the Figure 5 is exemplary; other types of structures may also be used to supplement or replace this structure to implement telecommunication functions or other functions.

[0130] In one or more embodiments of the present application, the search ranking function may be integrated into the central processing unit 9100. The central processing unit 9100 may be configured to perform the following control:

[0131] Step 100: Receive a target search statement.

[0132] Step 200: Obtain multiple recall statements from a preset operation and maintenance knowledge base according to the target search statement.

[0133] Step 300: Determine the operation and maintenance knowledge words corresponding to the target search statement and each recall statement, as well as the word weight and word vector of each operation and maintenance knowledge word, based on the pre-stored operation and maintenance knowledge graph, the word vector set, the target search statement and each recall statement.

[0134] Step 400: Determine and output the ranking results of the multiple recall statements based on the word weight and word vector of each operation and maintenance knowledge word.

[0135] It can be seen from the above description that the electronic device provided by the embodiments of the present application can improve the efficiency and accuracy of search sorting, thereby improving the reliability of data center operation and maintenance.

[0136] In another embodiment, the search ranking device can be configured separately from the central processing unit 9100. For example, the search ranking device can be configured as a chip connected to the central processing unit 9100, and the search ranking function is realized through the control of the central processing unit.

[0137] like Figure 5 As shown, the electronic device 9600 may further include: a communication module 9110, an input unit 9120, an audio processor 9130, a display 9160, and a power supply 9170. It is worth noting that the electronic device 9600 does not necessarily have to include Figure 5 In addition, the electronic device 9600 may also include all components shown in Figure 5 For components not shown, reference may be made to the prior art.

[0138] like Figure 5 As shown, the central processing unit 9100 is sometimes also referred to as a controller or operation control, and may include a microprocessor or other processor device and / or logic device. The central processing unit 9100 receives input and controls the operation of various components of the electronic device 9600.

[0139] Memory 9140 can be, for example, one or more of a cache, flash memory, hard drive, removable media, volatile memory, non-volatile memory, or other suitable devices. It can store the aforementioned failure-related information and also store programs that execute the relevant information. The CPU 9100 can execute the programs stored in memory 9140 to implement information storage or processing.

[0140] The input unit 9120 provides input to the central processing unit 9100. The input unit 9120 may be, for example, a keypad or touch input device. The power supply 9170 is used to provide power to the electronic device 9600. The display 9160 is used to display objects such as images and text. The display may be, for example, an LCD display, but is not limited thereto.

[0141] The memory 9140 may be a solid-state memory, such as a read-only memory (ROM), a random access memory (RAM), or a SIM card. Alternatively, it may be a memory that retains information even when power is off, can be selectively erased, and is provided with more data. Examples of such memory are sometimes referred to as EPROMs. The memory 9140 may also be some other type of device. The memory 9140 includes a buffer memory 9141 (sometimes referred to as a buffer). The memory 9140 may include an application / function storage unit 9142 for storing application programs and function programs or processes for executing the operation of the electronic device 9600 by the central processing unit 9100.

[0142] The memory 9140 may also include a data storage unit 9143 for storing data, such as contacts, digital data, pictures, sounds, and / or any other data used by the electronic device. The driver storage unit 9144 of the memory 9140 may include various driver programs for communication functions of the electronic device and / or for executing other functions of the electronic device (such as messaging applications, address book applications, etc.).

[0143] The communication module 9110 is a transmitter / receiver 9110 that transmits and receives signals via an antenna 9111. The communication module (transmitter / receiver) 9110 is coupled to the central processor 9100 to provide input signals and receive output signals, which may be the same as in a conventional mobile communication terminal.

[0144] Based on different communication technologies, multiple communication modules 9110 can be provided in the same electronic device, such as a cellular network module, a Bluetooth module, and / or a wireless local area network module. The communication module (transmitter / receiver) 9110 is also coupled to a speaker 9131 and a microphone 9132 via an audio processor 9130 to provide audio output via the speaker 9131 and receive audio input from the microphone 9132, thereby implementing common telecommunication functions. The audio processor 9130 may include any suitable buffer, decoder, amplifier, etc. Furthermore, the audio processor 9130 is also coupled to the central processing unit 9100, enabling local recording via the microphone 9132 and playback of stored audio via the speaker 9131.

[0145] As can be seen from the above description, the electronic device provided by the embodiments of the present application can improve the efficiency and accuracy of search sorting, thereby improving the reliability of data center operation and maintenance.

[0146] The embodiments of the present application also provide a computer-readable storage medium capable of implementing all steps of the search ranking method in the above embodiments. The computer-readable storage medium stores a computer program. When the computer program is executed by a processor, all steps of the search ranking method in the above embodiments are implemented. For example, when the processor executes the computer program, the following steps are implemented:

[0147] Step 100: Receive a target search statement.

[0148] Step 200: Obtain multiple recall statements from a preset operation and maintenance knowledge base according to the target search statement.

[0149] Step 300: Determine the operation and maintenance knowledge words corresponding to the target search statement and each recall statement, as well as the word weight and word vector of each operation and maintenance knowledge word, based on the pre-stored operation and maintenance knowledge graph, the word vector set, the target search statement and each recall statement.

[0150] Step 400: Determine and output the ranking results of the multiple recall statements based on the word weight and word vector of each operation and maintenance knowledge word.

[0151] It can be seen from the above description that the computer-readable storage medium provided in the embodiments of the present application can improve the efficiency and accuracy of search sorting, thereby improving the reliability of data center operation and maintenance.

[0152] In this application, the various embodiments of the above method are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments. For related parts, please refer to the partial description of the method embodiment.

[0153] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0154] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the steps in the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0155] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0156] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0157] Specific embodiments are used in this application to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A search ranking method, characterized in that: include: receiving a target search statement; Obtaining multiple recall statements from a preset operation and maintenance knowledge base according to the target search statement; Determine the operation and maintenance knowledge words corresponding to the target search statement and each recall statement, as well as the word weight and word vector of each operation and maintenance knowledge word, based on the pre-stored operation and maintenance knowledge graph, the word vector set, the target search statement, and each recall statement; Determine and output the sorting results of the multiple recall statements based on the word weight and word vector of each operation and maintenance knowledge word; The step of determining the operation and maintenance knowledge words corresponding to the target search statement and each recall statement, as well as the word weight and word vector of each operation and maintenance knowledge word, based on the pre-stored operation and maintenance knowledge graph, the word vector set, the target search statement, and each recall statement, includes: Determining the operation and maintenance knowledge words corresponding to the target search statement and each recall statement, as well as the word weight of each operation and maintenance knowledge word, based on the pre-stored operation and maintenance knowledge graph, the target search statement, and each recall statement; Determine the word vector for each operation and maintenance knowledge word based on the pre-stored word vector set; The step of determining the operation and maintenance knowledge words corresponding to the target search statement and each recall statement and the word weight of each operation and maintenance knowledge word according to the pre-stored operation and maintenance knowledge graph, the target search statement, and each recall statement includes: Perform word segmentation on the target search statement to obtain the operation and maintenance knowledge word corresponding to the target search statement, and determine whether the operation and maintenance knowledge word exists in the pre-stored operation and maintenance knowledge graph. If so, obtain the word weight of the operation and maintenance knowledge word from the pre-stored operation and maintenance knowledge graph; otherwise, set the word weight of the operation and maintenance knowledge word corresponding to the target search statement to the inverse of the number of nodes in the graph.

2. The search ranking method according to claim 1, characterized in that: The step of determining and outputting the sorting results of the plurality of recall statements based on the word weight and word vector of each operation and maintenance knowledge word includes: Calculate similarity based on the word weights and word vectors of the operation and maintenance knowledge words in the target search sentence and the word weights and word vectors of the operation and maintenance knowledge words in each recall sentence; According to the similarity calculation result of each recall sentence, the ranking result of the plurality of recall sentences is determined and output.

3. The search ranking method according to claim 1, wherein: Before obtaining a plurality of recall statements from a preset operation and maintenance knowledge base according to the target search statement, the method further includes: Obtain original operation and maintenance knowledge documents; Storing the plurality of operation and maintenance knowledge statements in the original operation and maintenance knowledge document in the preset operation and maintenance knowledge base; The recall statement is an operation and maintenance knowledge statement corresponding to the target search statement obtained from the preset operation and maintenance knowledge base according to a recall algorithm.

4. The search ranking method according to claim 1, wherein: The step of obtaining multiple recall statements from a preset operation and maintenance knowledge base according to the target search statement includes: A plurality of recall statements are obtained from a preset operation and maintenance knowledge base according to a multi-way recall algorithm and the target search statement. The multi-way recall algorithm includes: TFIDF and BM25 algorithms.

5. The search ranking method according to claim 1, wherein: Before obtaining the word weight and word vector of each operation and maintenance knowledge word from the pre-stored operation and maintenance knowledge graph and word vector set, the method further includes: Obtaining an original operation and maintenance knowledge document, where the original operation and maintenance knowledge document includes multiple operation and maintenance knowledge statements; Performing word segmentation processing on the original operation and maintenance knowledge document to obtain a plurality of original operation and maintenance knowledge words; Applying the TextRank algorithm and the plurality of original operation and maintenance knowledge words to determine the word weight of each original operation and maintenance knowledge word; The operation and maintenance knowledge graph is obtained based on each original operation and maintenance knowledge word and its respective word weight, wherein the operation and maintenance knowledge graph includes: nodes, node weights and directed edges between nodes, nodes represent original operation and maintenance knowledge words, and directed edges represent the contextual relationship between original operation and maintenance knowledge words in operation and maintenance knowledge sentences.

6. The search ranking method according to claim 5, characterized in that: After performing word segmentation processing on the original operation and maintenance knowledge document to obtain a plurality of original operation and maintenance knowledge words, the method further includes: Perform embedding vectorization processing on each original operation and maintenance knowledge word to obtain the word vector of each original operation and maintenance knowledge word; The word vector set is obtained according to each original operation and maintenance knowledge word and its respective word vector.

7. A search and sorting device, characterized in that: include: A receiving module, configured to receive a target search statement; A recall module, configured to obtain a plurality of recall statements from a preset operation and maintenance knowledge base according to the target search statement; A determination module is used to determine the operation and maintenance knowledge words corresponding to the target search statement and each recall statement, as well as the word weight and word vector of each operation and maintenance knowledge word based on the pre-stored operation and maintenance knowledge graph, the word vector set, the target search statement and each recall statement; A sorting module, configured to determine and output sorting results of the plurality of recall statements based on the word weight and word vector of each operation and maintenance knowledge word; The step of determining the operation and maintenance knowledge words corresponding to the target search statement and each recall statement, as well as the word weight and word vector of each operation and maintenance knowledge word, based on the pre-stored operation and maintenance knowledge graph, the word vector set, the target search statement, and each recall statement, includes: Determining the operation and maintenance knowledge words corresponding to the target search statement and each recall statement, as well as the word weight of each operation and maintenance knowledge word, based on the pre-stored operation and maintenance knowledge graph, the target search statement, and each recall statement; Determine the word vector for each operation and maintenance knowledge word based on the pre-stored word vector set; The step of determining the operation and maintenance knowledge words corresponding to the target search statement and each recall statement and the word weight of each operation and maintenance knowledge word according to the pre-stored operation and maintenance knowledge graph, the target search statement, and each recall statement includes: Perform word segmentation on the target search statement to obtain the operation and maintenance knowledge word corresponding to the target search statement, and determine whether the operation and maintenance knowledge word exists in the pre-stored operation and maintenance knowledge graph. If so, obtain the word weight of the operation and maintenance knowledge word from the pre-stored operation and maintenance knowledge graph; otherwise, set the word weight of the operation and maintenance knowledge word corresponding to the target search statement to the inverse of the number of nodes in the graph.

8. The search ranking device according to claim 7, characterized in that: The sorting module includes: A scoring unit, configured to calculate similarity based on the word weights and word vectors of the operation and maintenance knowledge words of the target search sentence and the word weights and word vectors of the operation and maintenance knowledge words of each recall sentence; The determining unit is configured to determine and output a ranking result of the plurality of recall statements based on a similarity calculation result of each recall statement.

9. The search ranking device according to claim 7, characterized in that: Also includes: Acquisition module, used to obtain original operation and maintenance knowledge documents; A storage module, configured to store the plurality of operation and maintenance knowledge statements in the original operation and maintenance knowledge document in the preset operation and maintenance knowledge base; The recall statement is an operation and maintenance knowledge statement corresponding to the target search statement obtained from the preset operation and maintenance knowledge base according to a recall algorithm.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the search and sorting method according to any one of claims 1 to 6 is implemented.

11. A computer-readable storage medium having computer instructions stored thereon, characterized in that: When the instructions are executed, the search ranking method according to any one of claims 1 to 6 is implemented.

Images