CN117708270A - Enterprise data query method, device, equipment and storage medium - Google Patents
Enterprise data query method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN117708270A CN117708270A CN202311694539.7A CN202311694539A CN117708270A CN 117708270 A CN117708270 A CN 117708270A CN 202311694539 A CN202311694539 A CN 202311694539A CN 117708270 A CN117708270 A CN 117708270A
- Authority
- CN
- China
- Prior art keywords
- query
- data
- enterprise
- information
- index
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 60
- 238000012545 processing Methods 0.000 claims abstract description 14
- 239000011159 matrix material Substances 0.000 claims description 46
- 230000006399 behavior Effects 0.000 claims description 27
- 230000000977 initiatory effect Effects 0.000 claims description 21
- 238000007781 pre-processing Methods 0.000 claims description 8
- 238000001914 filtration Methods 0.000 claims description 6
- 238000012937 correction Methods 0.000 claims description 5
- 230000006870 function Effects 0.000 description 7
- 238000004891 communication Methods 0.000 description 6
- 238000005457 optimization Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 238000010276 construction Methods 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 206010063385 Intellectualisation Diseases 0.000 description 1
- 230000003542 behavioural effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000010485 coping Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000011478 gradient descent method Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000007115 recruitment Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 230000017105 transposition Effects 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses an enterprise data query method, device, equipment and storage medium, relating to the field of data processing, wherein the method comprises the following steps: responding to a query request, acquiring query information in the query request, searching a pre-constructed index library based on the query information, determining a target index, querying an enterprise knowledge graph, determining target event data matched with the target index in the enterprise knowledge graph, wherein the enterprise knowledge graph is constructed based on the index library in advance, and feeding back the query request based on the target event data; according to the method and the device for searching the enterprise data, the target index related to the query information is determined through searching the index library, the target event data matched with the target index in the enterprise knowledge graph is queried, and the query request is fed back based on the target event data, so that the enterprise data searching efficiency is effectively improved, the timeliness and the accuracy of enterprise data searching are ensured, and the enterprise management efficiency is effectively improved.
Description
Technical Field
The present invention relates to the field of data processing technologies, and in particular, to a method, an apparatus, a device, and a storage medium for querying enterprise data.
Background
With the development of technology, enterprise data is gradually increased, and due to huge and complex data information volume in enterprises, the timeliness and accuracy of search information are required to be correspondingly guaranteed, and enterprise business management can help enterprise managers or business personnel to improve the working efficiency, but not increase the burden of the enterprise managers or the business personnel. Thus, enterprise management software is focusing on the comprehensiveness of functions, the controllability of processes, the advancement of technology, and the ease of use of systems. Because the data information volume in enterprises is huge and complex, the timeliness and the accuracy of enterprise data searching cannot be guaranteed at present, and the enterprise business management efficiency is low.
The foregoing is provided merely for the purpose of facilitating understanding of the technical solutions of the present invention and is not intended to represent an admission that the foregoing is prior art.
Disclosure of Invention
The invention mainly aims to provide an enterprise data query method, device, equipment and storage medium, and aims to solve the technical problem that the efficiency of enterprise business management is low because the timeliness and the accuracy of enterprise data search cannot be guaranteed in the prior art.
To achieve the above object, the present invention provides an enterprise data query method, the method comprising the steps of:
responding to a query request, and acquiring query information in the query request;
searching a pre-constructed index library based on the query information, and determining a target index;
querying an enterprise knowledge graph, and determining target event data matched with the target index in the enterprise knowledge graph, wherein the enterprise knowledge graph is constructed in advance based on the index library;
and feeding back the query request based on the target event data.
Optionally, the searching the pre-constructed index library based on the query information, before determining the target index, further includes:
extracting features of the original data to obtain data feature information;
carrying out service classification on the original data based on the data characteristic information, and determining associated data associated with each field in the original data based on a service classification result;
determining a similarity index between the associated data;
and constructing an index library corresponding to the original data based on the similarity index.
Optionally, the feeding back the query request based on the target event data includes:
acquiring account matrix positioning information of an initiating user of the query request;
constructing an account frequency matrix based on the account matrix positioning information, and constructing a data packet content matrix;
carrying out data preference prediction based on the account frequency matrix and the data packet content matrix to obtain a prediction result;
and generating an enterprise data packet based on the target event data and the prediction result, and feeding back the query request based on the enterprise data packet.
Optionally, the querying the enterprise knowledge graph, determining target event data matched with the target index in the enterprise knowledge graph includes:
querying an enterprise knowledge graph to determine at least one candidate event related to the target index in the enterprise knowledge graph;
determining the degree of correlation between each candidate event and the target index, and extracting event characteristic information of each candidate event;
performing weight configuration on each candidate event based on the correlation degree and the event characteristic information;
and determining target event data matched with the target index in the candidates based on the result of the weight configuration.
Optionally, the searching the pre-constructed index base based on the query information to determine the target index includes:
preprocessing the query information to obtain a query corpus;
extracting search keywords from the query corpus;
searching a pre-constructed index library based on the search keywords, and determining a target index.
Optionally, the preprocessing the query information to obtain a query corpus includes:
extracting corpus from the query information to obtain initial corpus;
filtering sensitive words of the initial corpus to obtain a candidate corpus;
and carrying out error correction processing on the candidate corpus to obtain query corpus.
Optionally, the extracting the search keyword from the query corpus includes:
acquiring historical behavior information of an initiating user of the query request;
performing behavior analysis on the historical behavior information to obtain behavior characteristic information of the initiating user;
performing context analysis on the query corpus to obtain query context information;
determining search intention information of the initiating user based on the behavior feature information and the query context information;
and extracting search keywords from the query corpus based on the search intention information.
In addition, to achieve the above object, the present invention also proposes an enterprise data query apparatus, including:
the information acquisition module is used for responding to the query request and acquiring the query information in the query request;
the index searching module is used for searching a pre-constructed index library based on the query information and determining a target index;
the map query module is used for querying an enterprise knowledge map, determining target event data matched with the target index in the enterprise knowledge map, and constructing the enterprise knowledge map on the basis of the index library in advance;
and the query feedback module is used for feeding back the query request based on the target event data.
In addition, to achieve the above object, the present invention also proposes an enterprise data query apparatus, including: a memory, a processor, and an enterprise data query program stored on the memory and executable on the processor, the enterprise data query program configured to implement the steps of the enterprise data query method as described above.
In addition, to achieve the above object, the present invention also proposes a storage medium having stored thereon an enterprise data query program which, when executed by a processor, implements the steps of the enterprise data query method as described above.
According to the method, query information in a query request is obtained by responding to the query request, a pre-built index library is searched based on the query information, a target index is determined, an enterprise knowledge graph is queried, target event data matched with the target index in the enterprise knowledge graph is determined, the enterprise knowledge graph is pre-built based on the index library, and the query request is fed back based on the target event data; the invention searches the index library, determines the target index related to the query information, queries the target event data matched with the target index in the enterprise knowledge graph, and feeds back the query request based on the target event data.
Drawings
FIG. 1 is a schematic diagram of an enterprise data query device of a hardware operating environment in accordance with an embodiment of the present invention;
FIG. 2 is a flowchart of a first embodiment of an enterprise data query method according to the present invention;
FIG. 3 is a flowchart of a second embodiment of an enterprise data query method according to the present invention;
FIG. 4 is a flowchart of a third embodiment of an enterprise data query method according to the present invention;
fig. 5 is a block diagram of a first embodiment of an enterprise data query apparatus in accordance with the present invention.
The achievement of the objects, functional features and advantages of the present invention will be further described with reference to the accompanying drawings, in conjunction with the embodiments.
Detailed Description
It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
Referring to fig. 1, fig. 1 is a schematic structural diagram of an enterprise data query device in a hardware running environment according to an embodiment of the present invention.
As shown in fig. 1, the enterprise data query apparatus may include: a processor 1001, such as a central processing unit (Central Processing Unit, CPU), a communication bus 1002, a user interface 1003, a network interface 1004, a memory 1005. Wherein the communication bus 1002 is used to enable connected communication between these components. The user interface 1003 may include a Display, an input unit such as a Keyboard (Keyboard), and the optional user interface 1003 may further include a standard wired interface, a wireless interface. The network interface 1004 may optionally include a standard wired interface, a Wireless interface (e.g., a Wireless-Fidelity (Wi-Fi) interface). The Memory 1005 may be a high-speed random access Memory (Random Access Memory, RAM) or a stable nonvolatile Memory (NVM), such as a disk Memory. The memory 1005 may also optionally be a storage device separate from the processor 1001 described above.
Those skilled in the art will appreciate that the structure shown in FIG. 1 is not limiting of the enterprise data query device, and may include more or fewer components than shown, or may combine certain components, or a different arrangement of components.
As shown in fig. 1, an operating system, a network communication module, a user interface module, and an enterprise data query program may be included in the memory 1005 as one type of storage medium.
In the enterprise data querying device shown in fig. 1, the network interface 1004 is mainly used for data communication with a network server; the user interface 1003 is mainly used for data interaction with a user; the processor 1001 and the memory 1005 in the enterprise data query apparatus of the present invention may be disposed in the enterprise data query apparatus, and the enterprise data query apparatus invokes an enterprise data query program stored in the memory 1005 through the processor 1001 and executes the enterprise data query method provided by the embodiment of the present invention.
An embodiment of the present invention provides an enterprise data query method, and referring to fig. 2, fig. 2 is a flow chart of a first embodiment of an enterprise data query method according to the present invention.
In this embodiment, the enterprise data query method includes the following steps:
step S10: and responding to the query request, and acquiring query information in the query request.
It should be noted that, the enterprise search is to index various structured and unstructured information inside the enterprise through the search software, and provide a search method. In enterprises, managers often assist the manager in making various decisions and coping with various emergencies by searching related information of the enterprises, and due to huge and complex data information quantity in the enterprises, the timeliness and the accuracy of the searched information are required to be correspondingly ensured.
Because the data information volume in enterprises is huge and complex, the timeliness and the accuracy of enterprise information search cannot be ensured at present, and the enterprise business management efficiency is low. The method and the device are applied to determining the target index related to query information by searching the index library when a query request of enterprise data is received, querying target event data matched with the target index in the enterprise knowledge graph, and feeding back the query request based on the target event data.
It should be understood that the main body of execution of the method of this embodiment may be an enterprise data query device with functions of data processing, network communication and program running, such as a computer, or other apparatus or device capable of implementing the same or similar functions, where the above enterprise data query device (hereinafter referred to as a data query device) is described as an example.
It should be noted that, the query request may be an enterprise data query request input by a user, that is, a query instruction input by the user. The query information may be query keyword information and/or target query content information extracted from the data query request by the data query device.
It can be understood that the data query device extracts and obtains the query keyword information and/or the target query content information by preprocessing the query request input by the user and analyzing the preprocessed query request.
Step S20: searching a pre-constructed index library based on the query information, and determining a target index.
It can be appreciated that the data query device searches the index library constructed in advance based on the query information to obtain at least one index result, sorts the index results based on the relevance and the current search scene by analyzing the relevance of each index result, and determines the target index based on the sorted results.
Further, in order to accurately retrieve the target index related to the query information, the step S20 may include:
preprocessing the query information to obtain a query corpus;
extracting search keywords from the query corpus;
searching a pre-constructed index library based on the search keywords, and determining a target index.
The preprocessing may be an analytical processing such as data cleansing for the query information. The query corpus may be a text data set obtained by extracting the query information after analysis and processing. The keywords may be keywords related to data query and/or data retrieval in a query corpus, and the like.
It will be appreciated that the present embodiment may construct an index base based on an inverted index, which is a data structure that represents a mapping that indexes words or numbers as keys to all documents or database files in which the words or words occur. The inverted index is composed of three parts, a keyword, a term dictionary, and a posting list.
In a specific implementation, the indexing process first needs to find the location of the term index. term index is used to find the location of the keyword term in term dictionary. terminax has a wide variety of dictionary structures such as hash tables, B-trees, b+, FST.
As an example, for example, if there is only one sentence "high frequency refresh rate display" in a text, before the index is created, the index table is split and formed according to the word splitter, and if the index table is split into three terms according to the word splitter, the high frequency refresh rate display and the high frequency refresh rate display are respectively. Creating the inverted index table is to record the information content such as the number of Times (TF) these term words appear in the document, which locations in the document the words are located, etc.
Further, in order to effectively process the query information, the preprocessing the query information to obtain the query corpus may include:
extracting corpus from the query information to obtain initial corpus;
filtering sensitive words of the initial corpus to obtain a candidate corpus;
and carrying out error correction processing on the candidate corpus to obtain query corpus.
It may be appreciated that the data query device may identify the user search intent by performing an analysis process on the initial corpus extracted from the query information, which may include, for example, performing sensitive word filtering, error correction, synonym processing, etc. on the initial corpus, and then initiating a historical behavioral analysis, context analysis, etc. of the user based on the query request.
Further, in order to accurately extract the keywords in the query corpus, the extracting the search keywords from the query corpus may include:
acquiring historical behavior information of an initiating user of the query request;
performing behavior analysis on the historical behavior information to obtain behavior characteristic information of the initiating user;
performing context analysis on the query corpus to obtain query context information;
determining search intention information of the initiating user based on the behavior feature information and the query context information;
and extracting search keywords from the query corpus based on the search intention information.
It can be understood that the data query device constructs a behavior representation corresponding to the initiating user by acquiring historical behavior information of the initiating user of the query request, performing behavior analysis on the historical behavior information, acquiring behavior feature information of the initiating user based on the behavior representation, performing context analysis on the query corpus, determining context in the corpus, obtaining query context information based on the context analysis, determining search intention information of the initiating user based on the behavior feature information and the query context information, segmenting the query corpus based on the search intention information, and extracting search keywords based on the segmentation result.
Step S30: querying an enterprise knowledge graph, and determining target event data matched with the target index in the enterprise knowledge graph.
It should be noted that, the enterprise knowledge graph is constructed in advance based on the index base. The data query device integrates various enterprise data in advance to build a knowledge graph database, wherein the various enterprise data comprises but is not limited to: enterprise base data, investment relationships, job relationships, enterprise patent data, enterprise bid data, enterprise recruitment data, enterprise litigation data, enterprise trust loss data, enterprise news data, and the like.
It can be appreciated that the data query device automatically builds a structured tag set for entity portraits in the enterprise knowledge graph by analyzing RDF data in the knowledge graph database; the entity portraits are matched and marked through the labels in the structured label set, so that a user can intuitively and rapidly distinguish and compare the entities in the knowledge graph through the entity portrait result; an initial knowledge-graph is constructed based on the plurality of structured tag sets. And adding a retrieval index in an index library into the constructed initial knowledge graph to obtain an enterprise knowledge graph.
Further, in order to accurately retrieve the target event data matching the target index, the step S30 may include:
querying an enterprise knowledge graph to determine at least one candidate event related to the target index in the enterprise knowledge graph;
determining the degree of correlation between each candidate event and the target index, and extracting event characteristic information of each candidate event;
performing weight configuration on each candidate event based on the correlation degree and the event characteristic information;
and determining target event data matched with the target index in the candidates based on the result of the weight configuration.
It should be noted that, the candidate event may be an event associated with the target index in the enterprise knowledge graph, and the candidate event may be an object of the knowledge in the enterprise knowledge graph, that is, the object being indexed.
It can be understood that the data query device determines at least one candidate event related to the target index in the enterprise knowledge graph by querying the enterprise knowledge graph, if a plurality of candidate events exist, the data query device ranks the relevance of each candidate event according to the index content, the ranking can perform individual ranking according to different scenes and user information of a data query request initiating user, weight configuration is performed on each candidate event based on a ranking result, and target event data with higher relevance is screened from the candidate events based on a weight configuration result.
Step S40: and feeding back the query request based on the target event data.
It should be noted that, the event may be the content of knowledge in the enterprise knowledge graph, that is, the indexed object, and the target event data, that is, the data related to the event content.
It can be understood that, in this embodiment, the query request is fed back based on the target event data obtained by searching in the enterprise knowledge graph, and the search result is output.
In a specific implementation, the data query device builds an enterprise knowledge graph of enterprise-level search in advance, adds a search index into the built enterprise knowledge graph, performs knowledge search based on the enterprise knowledge graph, and determines the shortest path between events. Wherein, the event represents knowledge content of each business field in the enterprise knowledge graph; and constructing a training set based on the shortest path, constructing an event search model based on the training set, and completing event retrieval shortest path simulation based on a deep learning algorithm. The event search model refers to a search processing model of event content in each field, the event search model in each field needs large-scale training and labeling, and the event search model is used for constructing and storing search data indexes; the simulation is used for obtaining the searched paths of all events and obtaining the shortest path through algorithm comparison.
According to the method, query information in a query request is obtained by responding to the query request, a pre-built index base is searched based on the query information, a target index is determined, an enterprise knowledge graph is queried, target event data matched with the target index in the enterprise knowledge graph is determined, the enterprise knowledge graph is pre-built based on the index base, and the query request is fed back based on the target event data; according to the method and the device, the index library is searched, the target index related to the query information is determined, the target event data matched with the target index in the enterprise knowledge graph is queried, and the query request is fed back based on the target event data.
Referring to fig. 3, fig. 3 is a flowchart of a second embodiment of an enterprise data query method according to the present invention.
Based on the first embodiment, in this embodiment, before step S20, the method includes:
step S21: extracting features of the original data to obtain data feature information;
step S22: carrying out service classification on the original data based on the data characteristic information, and determining associated data associated with each field in the original data based on a service classification result;
step S23: determining a similarity index between the associated data;
step S24: and constructing an index library corresponding to the original data based on the similarity index.
It can be understood that the data characteristic information can be characteristic information such as data type, service type and the like of various types of original data. The fields can be different service fields, and because the data of the different service fields are different, the service classification needs to be performed on the original data, and the data corresponding to each field is associated to the corresponding service field, so that a high-performance index library is constructed.
It should be noted that, in this embodiment, a data optimization scheme is introduced to optimize the storage of the index. The program data required by the service field for executing the service are required to be stored in the corresponding memory after being acquired, and the program data are different when the service field is different, so that the data volume is more.
It can be understood that the introduction of the data optimization method according to this embodiment needs to be implemented according to the following steps:
a1, recording the characteristics of program data in an aggregate structure; the elements of the set are the characteristics of the program data, and the number of the elements is the characteristic number of the program data;
a2, calculating a similarity index Q between different program data in the same service field, wherein the similarity index is calculated by referring to the following formula 1:
wherein the set A, B represents a feature set of two different program data in the same service area, card (a n B) represents a common feature of two different program data in the same service area, and Card (a n B) represents a total number of different features in the feature set of two different program data; u (x) represents a filter function for filtering invalid data, referring to the following formula 2.
Wherein x=card (a) or Card (B);
a3, establishing an index between two program data with the similarity index Q larger than a preset threshold k, so that when one of the two program data is called, the system can call the other program data in time, the retrieval capability of the program data is greatly enhanced, and therefore optimization is achieved, and efficiency is improved.
In some embodiments, for example, one text data field is mapped to the program data set a {2, 5, 4}, another text data field is mapped to the program data set B {3, 5, 9}, x=3, u (x) =1, q=0.5 are calculated according to the formula, and if the threshold is 0.4, the document B can be returned after indexing to the document a, wherein the threshold requires a lot of data to be trained according to the business channels.
It should be noted that, in the data optimization method in this embodiment, an index is established between two program data with a similarity index Q greater than a preset threshold k through the similarity index Q, so that when one of the two program data is called, the system can call the other program data in time, and the retrieval capability of the program data is greatly enhanced, thereby realizing optimization and improving efficiency.
According to the method, the device and the system, the data characteristic information is obtained through characteristic extraction of the original data, the service classification is carried out on the original data based on the data characteristic information, the associated data associated with each field in the original data is determined based on the service classification result, the similarity index between the associated data is determined, and the index base corresponding to the original data is constructed based on the similarity index, so that the data retrieval efficiency is improved, and the storage optimization of index construction is achieved.
Referring to fig. 4, fig. 4 is a flowchart of a third embodiment of an enterprise data query method according to the present invention.
Based on the first embodiment, in this embodiment, the step S40 includes:
step S41: acquiring account matrix positioning information of an initiating user of the query request;
step S42: constructing an account frequency matrix based on the account matrix positioning information, and constructing a data packet content matrix;
step S43: carrying out data preference prediction based on the account frequency matrix and the data packet content matrix to obtain a prediction result;
step S44: and generating an enterprise data packet based on the target event data and the prediction result, and feeding back the query request based on the enterprise data packet.
Before the search result is output and fed back, in this embodiment, in order to improve the intellectualization and search efficiency of enterprise search, the enterprise data is packaged in the form of enterprise data packets, the content of each enterprise data packet is different, and after each user uses the data packet, the enterprise data packet is scored or not scored, where scoring refers to the use frequency score of the corresponding data packet when the service personnel logs in the account. The content of the enterprise data packet is extracted, such as form class, flow class, statistics class, etc.
It will be appreciated that the data querying device first establishes an account (i.e., the account corresponding to the initiating user of the data query request) use frequency matrix and a data packet content matrix. For the frequency matrix of the account use, firstly, the positioning of the account matrix is clarified, then, the matrix is built, for example, one account is in the image-text direction, and the layout in the four directions of video, audio, image-text and question-answering can be selected, and the four directions can be subdivided into different sub-accounts. For a data packet content matrix, the data content description includes data describing one or more files, each of which may be a computer readable file, wherein the matrix code is encoded with links to the one or more files, thereby converting domain knowledge into the data packet content matrix for ease of computation.
Then, the data query device performs matrix multiplication on the account use frequency matrix and the data packet content matrix to obtain a prediction result of business personnel on the enterprise data packet, wherein the prediction result comprises scoring prediction of business personnel on the unscored enterprise data packet, and the purpose is that the closer the prediction result is to the real situation, the better. Therefore, the square difference loss function is constructed by combining the predicted value with the value of the scored part in the account usage frequency table X, referring to the following formula 3:
wherein i represents the ith account, j represents the jth enterprise data packet, d represents the jth content, x represents the account preference matrix, w represents the data packet content matrix, y represents the usage frequency score matrix, r represents the score record matrix, no score is 0, and if so, 1, r (i, j) =1 represents that business person i has scored the enterprise data packet j, and r (i, j) =0 represents that business person i has not scored the enterprise data packet j.
The squared difference loss function is then minimized, and the correct parameters are obtained by machine learning using a gradient descent method to minimize the loss function.
For each parameter, a bias is obtained, referring to the following equations 4 and 5:
the gradient is as follows equation 6 and equation 7:
Δx＝r·(xw-y)w T equation 6
Δw＝x T [(xw-y)·r]Equation 7
Wherein, represents a point multiplication, and the non-rule represents a matrix multiplication, and the superscript T represents a matrix transposition.
And finally, carrying out gradient descent processing on the data query equipment to obtain correct parameters, and according to the parameters, realizing prediction recommendation, recommending enterprise data packets with high use frequency scores and favorites to corresponding business personnel, so as to realize efficient and intelligent enterprise searching.
In this embodiment, by adding prediction recommendation in the enterprise data search, the prediction recommendation can be implemented according to the parameters, and the enterprise data packets with high frequency of use scores and favorites are recommended to the corresponding service personnel, so that efficient and intelligent enterprise search can be implemented.
According to the method, the account frequency matrix is constructed based on the account matrix positioning information, the data packet content matrix is constructed, the data preference prediction is carried out based on the account frequency matrix and the data packet content matrix, the prediction result is obtained, the enterprise data packet is generated based on the target event data and the prediction result, and the query request is fed back based on the enterprise data packet, so that enterprise searching efficiency is effectively improved, intelligent searching is achieved, enterprise data searching is guaranteed to meet user requirements, and user personalized searching is achieved.
In addition, the embodiment of the invention also provides a storage medium, wherein the storage medium stores an enterprise data query program, and the enterprise data query program realizes the steps of the enterprise data query method when being executed by a processor.
Because the storage medium adopts all the technical solutions of all the embodiments, at least all the beneficial effects brought by the technical solutions of the embodiments are not described in detail herein.
Referring to fig. 5, fig. 5 is a block diagram illustrating a first embodiment of an enterprise data query apparatus in accordance with the present invention.
As shown in fig. 5, an enterprise data query apparatus according to an embodiment of the present invention includes:
an information acquisition module 10, configured to respond to a query request and acquire query information in the query request;
an index searching module 20, configured to search a pre-constructed index library based on the query information, and determine a target index;
the map query module 30 is configured to query an enterprise knowledge map, determine target event data matched with the target index in the enterprise knowledge map, where the enterprise knowledge map is previously constructed based on the index library;
and a query feedback module 40, configured to feed back the query request based on the target event data.
Further, the enterprise data query apparatus further includes:
the index construction module 50 is used for extracting features of the original data to obtain data feature information; carrying out service classification on the original data based on the data characteristic information, and determining associated data associated with each field in the original data based on a service classification result; determining a similarity index between the associated data; and constructing an index library corresponding to the original data based on the similarity index.
Further, the query feedback module 40 is further configured to obtain account matrix positioning information of the user who initiates the query request; constructing an account frequency matrix based on the account matrix positioning information, and constructing a data packet content matrix; carrying out data preference prediction based on the account frequency matrix and the data packet content matrix to obtain a prediction result; and generating an enterprise data packet based on the target event data and the prediction result, and feeding back the query request based on the enterprise data packet.
Further, the map query module 30 is further configured to query an enterprise knowledge map to determine at least one candidate event related to the target index in the enterprise knowledge map; determining the degree of correlation between each candidate event and the target index, and extracting event characteristic information of each candidate event; performing weight configuration on each candidate event based on the correlation degree and the event characteristic information; and determining target event data matched with the target index in the candidates based on the result of the weight configuration.
Further, the index search module 20 is further configured to pre-process the query information to obtain a query corpus; extracting search keywords from the query corpus; searching a pre-constructed index library based on the search keywords, and determining a target index.
Further, the index search module 20 is further configured to extract a corpus from the query information, and obtain an initial corpus; filtering sensitive words of the initial corpus to obtain a candidate corpus; and carrying out error correction processing on the candidate corpus to obtain query corpus.
Further, the index search module 20 is further configured to obtain historical behavior information of the user who initiates the query request; performing behavior analysis on the historical behavior information to obtain behavior characteristic information of the initiating user; performing context analysis on the query corpus to obtain query context information; determining search intention information of the initiating user based on the behavior feature information and the query context information; and extracting search keywords from the query corpus based on the search intention information.
According to the method, query information in a query request is obtained by responding to the query request, a pre-built index base is searched based on the query information, a target index is determined, an enterprise knowledge graph is queried, target event data matched with the target index in the enterprise knowledge graph is determined, the enterprise knowledge graph is pre-built based on the index base, and the query request is fed back based on the target event data; according to the method and the device, the index library is searched, the target index related to the query information is determined, the target event data matched with the target index in the enterprise knowledge graph is queried, and the query request is fed back based on the target event data.
It should be understood that the foregoing is illustrative only and is not limiting, and that in specific applications, those skilled in the art may set the invention as desired, and the invention is not limited thereto.
It should be noted that the above-described working procedure is merely illustrative, and does not limit the scope of the present invention, and in practical application, a person skilled in the art may select part or all of them according to actual needs to achieve the purpose of the embodiment, which is not limited herein.
In addition, technical details that are not described in detail in this embodiment may refer to the enterprise data query method provided in any embodiment of the present invention, which is not described herein again.
Furthermore, it should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or system. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or system that comprises the element.
The foregoing embodiment numbers of the present invention are merely for the purpose of description, and do not represent the advantages or disadvantages of the embodiments.
From the above description of the embodiments, it will be clear to those skilled in the art that the above-described embodiment method may be implemented by means of software plus a necessary general hardware platform, but of course may also be implemented by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium (e.g. Read Only Memory)/RAM, magnetic disk, optical disk) and including several instructions for causing a terminal device (which may be a mobile phone, a computer, a server, or a network device, etc.) to perform the method according to the embodiments of the present invention.
The foregoing description is only of the preferred embodiments of the present invention, and is not intended to limit the scope of the invention, but rather is intended to cover any equivalents of the structures or equivalent processes disclosed herein or in the alternative, which may be employed directly or indirectly in other related arts.
Claims (10)
1. An enterprise data query method, characterized in that the enterprise data query method comprises:
responding to a query request, and acquiring query information in the query request;
searching a pre-constructed index library based on the query information, and determining a target index;
querying an enterprise knowledge graph, and determining target event data matched with the target index in the enterprise knowledge graph, wherein the enterprise knowledge graph is constructed in advance based on the index library;
and feeding back the query request based on the target event data.
2. The enterprise data query method of claim 1, wherein searching a pre-built index base based on the query information, prior to determining a target index, further comprises:
extracting features of the original data to obtain data feature information;
carrying out service classification on the original data based on the data characteristic information, and determining associated data associated with each field in the original data based on a service classification result;
determining a similarity index between the associated data;
and constructing an index library corresponding to the original data based on the similarity index.
3. The enterprise data query method of claim 1, wherein the feeding back the query request based on the target event data comprises:
acquiring account matrix positioning information of an initiating user of the query request;
constructing an account frequency matrix based on the account matrix positioning information, and constructing a data packet content matrix;
carrying out data preference prediction based on the account frequency matrix and the data packet content matrix to obtain a prediction result;
and generating an enterprise data packet based on the target event data and the prediction result, and feeding back the query request based on the enterprise data packet.
4. The enterprise data query method of claim 1, wherein the querying an enterprise knowledge-graph, determining target event data in the enterprise knowledge-graph that matches the target index, comprises:
querying an enterprise knowledge graph to determine at least one candidate event related to the target index in the enterprise knowledge graph;
determining the degree of correlation between each candidate event and the target index, and extracting event characteristic information of each candidate event;
performing weight configuration on each candidate event based on the correlation degree and the event characteristic information;
and determining target event data matched with the target index in the candidates based on the result of the weight configuration.
5. The enterprise data query method of any one of claims 1 to 4, wherein the searching a pre-built index base based on the query information, determining a target index, comprises:
preprocessing the query information to obtain a query corpus;
extracting search keywords from the query corpus;
searching a pre-constructed index library based on the search keywords, and determining a target index.
6. The enterprise data query method of claim 5, wherein preprocessing the query information to obtain query corpus comprises:
extracting corpus from the query information to obtain initial corpus;
filtering sensitive words of the initial corpus to obtain a candidate corpus;
and carrying out error correction processing on the candidate corpus to obtain query corpus.
7. The enterprise data query method of claim 6, wherein the extracting search keywords from the query corpus comprises:
acquiring historical behavior information of an initiating user of the query request;
performing behavior analysis on the historical behavior information to obtain behavior characteristic information of the initiating user;
performing context analysis on the query corpus to obtain query context information;
determining search intention information of the initiating user based on the behavior feature information and the query context information;
and extracting search keywords from the query corpus based on the search intention information.
8. An enterprise data query apparatus, the enterprise data query apparatus comprising:
the information acquisition module is used for responding to the query request and acquiring the query information in the query request;
the index searching module is used for searching a pre-constructed index library based on the query information and determining a target index;
the map query module is used for querying an enterprise knowledge map, determining target event data matched with the target index in the enterprise knowledge map, and constructing the enterprise knowledge map on the basis of the index library in advance;
and the query feedback module is used for feeding back the query request based on the target event data.
9. An enterprise data query device, the enterprise data query device comprising: a memory, a processor, and an enterprise data query program stored on the memory and executable on the processor, the enterprise data query program configured to implement the enterprise data query method of any of claims 1 to 7.
10. A storage medium having stored thereon an enterprise data query program, which when executed by a processor, implements the enterprise data query method of any of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311694539.7A CN117708270A (en) | 2023-12-11 | 2023-12-11 | Enterprise data query method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311694539.7A CN117708270A (en) | 2023-12-11 | 2023-12-11 | Enterprise data query method, device, equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117708270A true CN117708270A (en) | 2024-03-15 |
Family
ID=90161739
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311694539.7A Pending CN117708270A (en) | 2023-12-11 | 2023-12-11 | Enterprise data query method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117708270A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118014446A (en) * | 2024-04-09 | 2024-05-10 | 广东瑞和通数据科技有限公司 | Enterprise technology innovation comprehensive index analysis method, storage medium and computer equipment |
-
2023
- 2023-12-11 CN CN202311694539.7A patent/CN117708270A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118014446A (en) * | 2024-04-09 | 2024-05-10 | 广东瑞和通数据科技有限公司 | Enterprise technology innovation comprehensive index analysis method, storage medium and computer equipment |
| CN118014446B (en) * | 2024-04-09 | 2024-06-21 | 广东瑞和通数据科技有限公司 | Enterprise technology innovation comprehensive index analysis method, storage medium and computer equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111753060B (en) | Information retrieval method, apparatus, device and computer readable storage medium | |
| CN109829104B (en) | Semantic similarity based pseudo-correlation feedback model information retrieval method and system | |
| US8903810B2 (en) | Techniques for ranking search results | |
| CN111008265B (en) | Enterprise information searching method and device | |
| US8468156B2 (en) | Determining a geographic location relevant to a web page | |
| CN109885773B (en) | Personalized article recommendation method, system, medium and equipment | |
| US20040249808A1 (en) | Query expansion using query logs | |
| CN112035599B (en) | Query method and device based on vertical search, computer equipment and storage medium | |
| Nguyen et al. | A math-aware search engine for math question answering system | |
| US20030212663A1 (en) | Neural network feedback for enhancing text search | |
| CN112115232A (en) | Data error correction method and device and server | |
| US20220300543A1 (en) | Method of retrieving query, electronic device and medium | |
| CN115563313A (en) | Knowledge graph-based document book semantic retrieval system | |
| CN114722137A (en) | Security policy configuration method and device based on sensitive data identification and electronic equipment | |
| CN117708270A (en) | Enterprise data query method, device, equipment and storage medium | |
| US20120130999A1 (en) | Method and Apparatus for Searching Electronic Documents | |
| CN117539990A (en) | Problem processing method and device, electronic equipment and storage medium | |
| CN110727769A (en) | Corpus generation method and device, and man-machine interaction processing method and device | |
| WO2007124430A2 (en) | Search techniques using association graphs | |
| KR102256007B1 (en) | System and method for searching documents and providing an answer to a natural language question | |
| CN110688559A (en) | Retrieval method and device | |
| CN116414961A (en) | Question-answering method and system based on military domain knowledge graph | |
| CN111581326B (en) | Method for extracting answer information based on heterogeneous external knowledge source graph structure | |
| JP2010282403A (en) | Document retrieval method | |
| Kulkarni et al. | Information Retrieval based Improvising Search using Automatic Query Expansion |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |