Basin business document approximate search method and device, equipment and medium

Abstract

The invention relates to the technical field of data retrieval, and discloses a watershed business document approximate search method, device and equipment and a medium, and the method comprises the following steps: obtaining a watershed business knowledge document, and carrying out dimension reduction compression coding on the watershed business knowledge document to obtain a contrast value function; performing empirical mode decomposition on the contrast value function to obtain a plurality of intrinsic mode function components, and integrating the plurality of intrinsic mode function components in the same coordinate system; clustering the drainage basin business knowledge documents corresponding to the same intrinsic mode function component to obtain a plurality of clustered documents; obtaining a query document, respectively calculating mahalanobis distances between query points corresponding to the query document and the plurality of intrinsic mode function components, and selecting a clustering document as an approximate document based on the mahalanobis distances; according to the method, the document search efficiency and the response efficiency of watershed knowledge sharing are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the field of data retrieval technology, and in particular to a method, device, equipment and medium for approximate search of watershed business documents. Background Art

[0002] After deduplication, identification and sorting of watershed business data (including documents, etc.), watershed business knowledge is formed, and the demand for watershed knowledge sharing is derived. In this process, a large number of retrieval operations are generated, including data document search, related information search and other services.

[0003] However, the document similarity search, related information document search and other services of watershed business knowledge sharing have slow response speed and low search efficiency, which reduces the response efficiency of watershed knowledge sharing. Summary of the invention

[0004] In view of this, the present invention provides a method, device, equipment and medium for approximate search of watershed business documents to solve the problems of slow response speed and low search efficiency of watershed business knowledge document search service.

[0005] In a first aspect, the present invention provides a method for approximate search of a watershed business document, the method comprising: Acquire the watershed business knowledge document, perform dimension reduction compression coding on the watershed business knowledge document, and obtain the control value function; Performing empirical mode decomposition on the control value function to obtain several intrinsic mode function components, and integrating the several intrinsic mode function components in the same coordinate system; Cluster the watershed business knowledge documents corresponding to the same intrinsic mode function component respectively to obtain several cluster documents; The query document is obtained, the Mahalanobis distances between the query point corresponding to the query document and several intrinsic mode function components are calculated respectively, and the clustered documents are selected as approximate documents based on the Mahalanobis distances.

[0006] The present embodiment provides a method for approximate search of watershed business documents. By performing dimensionality reduction compression encoding on watershed business knowledge documents to obtain a comparison value function, the comparison value function is clustered by empirical mode decomposition, and similar documents are clustered, which reduces the number of calculations of document similarity. The Mahalanobis distance between a query point corresponding to a query document and several intrinsic mode function components is calculated, and clustered documents are selected as approximate documents based on the Mahalanobis distance. Only the clustered documents corresponding to the shortest Mahalanobis distance are obtained by comparison, which speeds up the data processing process, solves the problems of slow response speed and low search efficiency of services such as approximate search of documents for watershed business knowledge sharing and search of associated information documents, and significantly improves the document search efficiency and the response efficiency of watershed knowledge sharing.

[0007] In an optional implementation, the watershed business knowledge document is subjected to dimensionality reduction compression encoding to obtain a control value function, including: The keyword extraction algorithm is used to count the target keywords and the frequency of occurrence of the target keywords in each basin business knowledge document; The target keywords are weighted and encoded based on their occurrence frequency to obtain Huffman coding; A control value function is generated based on the target keyword, the occurrence frequency corresponding to the target keyword and the Huffman coding.

[0008] The present embodiment provides an approximate search method for watershed business documents, which utilizes a keyword extraction algorithm to respectively count the target keywords and the occurrence frequency of the target keywords of each watershed business knowledge document, and weights the target keywords according to the occurrence frequency of the target keywords to obtain Huffman coding, thereby realizing dimensionality reduction compression coding of the watershed business knowledge documents, mapping the watershed business knowledge documents into binary coding, and then generating a control value function based on the target keywords, the occurrence frequency corresponding to the target keywords and the Huffman coding, thereby laying a foundation for subsequent empirical mode decomposition clustering.

[0009] In an optional implementation, generating a control value function based on a target keyword, the occurrence frequency corresponding to the target keyword, and Huffman coding includes: Based on the target keyword, the frequency of occurrence of the target keyword and Huffman coding, the document comparison value corresponding to each watershed business knowledge document is calculated respectively; The document comparison values ​​are integrated to obtain the comparison value function.

[0010] This embodiment provides an approximate search method for watershed business documents, which calculates the document comparison value corresponding to each watershed business knowledge document, and then integrates the document comparison values ​​to obtain a comparison value function, thereby achieving accurate integration of document comparison values ​​and document identifiers.

[0011] In an optional implementation, before obtaining the watershed business knowledge document, performing dimension reduction compression encoding on the watershed business knowledge document, and obtaining the comparison value function, the method further includes: A unique document identifier is assigned to each watershed business knowledge document, and the document identifier is an integer not greater than the total number of documents.

[0012] In an optional implementation, empirical mode decomposition is performed on the control value function to obtain several intrinsic mode function components, including: Performing empirical mode decomposition on the control value function to obtain the sum of several intrinsic mode function components and residual terms, wherein the intrinsic mode function components include an upper envelope and a lower envelope; For each intrinsic mode function component, respectively calculate the mean of the upper envelope and the lower envelope corresponding to the intrinsic mode function component; The horizontal axis of the intrinsic mode function component is replaced by the mean value, and several intrinsic mode function components are integrated in the coordinate system.

[0013] The present embodiment provides an approximate search method for watershed business documents. By performing empirical mode decomposition on the control value function, the document clustering problem is converted into a signal decomposition problem, so that the characteristics of the signal at different time scales can be clearly identified and analyzed. Then, several intrinsic mode function components are integrated into the same coordinate system, which can provide a unified framework for subsequent similar document clustering and improve document search efficiency.

[0014] In an optional implementation, respectively calculating the Mahalanobis distances between the query point corresponding to the query document and a plurality of intrinsic mode function components, and selecting clustered documents as approximate documents based on the Mahalanobis distances, including: Calculate the Mahalanobis distances between the query point corresponding to the query document and several intrinsic mode function components respectively; The intrinsic modal function component corresponding to the minimum Mahalanobis distance is used as the target intrinsic modal function component, and the clustered documents corresponding to the target intrinsic modal function component are used as approximate documents.

[0015] This embodiment provides an approximate search method for watershed business documents. By calculating the Mahalanobis distance between a query point and several intrinsic modal function components, the document similarity problem is converted into a distance problem between a point and a distribution. This method can eliminate the influence of correlation between variables on distance calculation, accelerate the data processing process, realize efficient approximate search for similar documents, and reduce the number of calculations of document similarity.

[0016] In a second aspect, the present invention provides a device for approximate search of a watershed business document, the device comprising: A dimension reduction compression coding module is used to obtain a watershed business knowledge document, perform dimension reduction compression coding on the watershed business knowledge document, and obtain a comparison value function; An empirical mode decomposition module is used to perform empirical mode decomposition on the control value function to obtain a number of intrinsic mode function components, and the several intrinsic mode function components are integrated in the same coordinate system; A clustering module is used to cluster the watershed business knowledge documents corresponding to the same intrinsic mode function component to obtain a number of cluster documents; The calculation module is used to obtain the query document, calculate the Mahalanobis distance between the query point corresponding to the query document and several intrinsic mode function components, and select the cluster document as the approximate document based on the Mahalanobis distance.

[0017] In a third aspect, the present invention provides a computer device, comprising: a memory and a processor, the memory and the processor are communicatively connected to each other, computer instructions are stored in the memory, and the processor executes the approximate search method for watershed business documents of the first aspect or any corresponding embodiment thereof by executing the computer instructions.

[0018] In a fourth aspect, the present invention provides a computer-readable storage medium having computer instructions stored thereon, the computer instructions being used to enable a computer to execute the method for approximate search of watershed business documents according to the first aspect or any corresponding embodiment thereof.

[0019] In a fifth aspect, the present invention provides a computer program product, comprising computer instructions, wherein the computer instructions are used to enable a computer to execute the method for approximate search of watershed business documents according to the first aspect or any corresponding embodiment thereof. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0021] Figure 1 It is a flowchart of a method for approximate search of watershed business documents according to an embodiment of the present invention; Figure 2 is a flow chart of another method for approximate search of watershed business documents according to an embodiment of the present invention; Figure 3 is a flow chart of another method for approximate search of watershed business documents according to an embodiment of the present invention; Figure 4 is a structural block diagram of a device for approximate search of watershed business documents according to an embodiment of the present invention; Figure 5 It is a schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0022] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0023] An embodiment of the present invention provides an approximate search method for watershed business documents. For text documents of watershed business knowledge, a large number of documents with high similarity to the target documents can be effectively retrieved through approximate search, thereby reducing the number of calculations of document similarity and accelerating the data processing process, thereby significantly improving document search efficiency and improving watershed knowledge sharing response efficiency.

[0024] An embodiment of the present invention provides a method for approximate search of watershed business documents. It should be noted that the execution subject of the method for approximate search of watershed business documents provided by the embodiment of the present invention may be a device for approximate search of watershed business documents. The device for approximate search of watershed business documents may be implemented as part or all of an electronic device through software, hardware, or a combination of software and hardware. The electronic device may be a server or a terminal. The server in the embodiment of the present application may be a single server or a server cluster composed of multiple servers. The terminal in the embodiment of the present application may be a document search system server. The document search system server may be integrated in other intelligent hardware devices such as smart phones, personal computers, tablet computers, and intelligent robots. In the following method embodiments, the execution subject is an electronic device as an example for explanation.

[0025] According to an embodiment of the present invention, an embodiment of a method for approximate search of flow domain business documents is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0026] In this embodiment, a method for approximate search of flow domain business documents is provided, which can be used in the above-mentioned electronic device. Figure 1 is a flow chart of a method for approximate search of a flow domain business document according to an embodiment of the present invention. Figure 1 As shown, the process includes the following steps: Step S101, obtaining a watershed business knowledge document, performing dimensionality reduction compression coding on the watershed business knowledge document, and obtaining a control value function.

[0027] Specifically, the watershed business knowledge document contains watershed runoff data, watershed geographic data, watershed climate data, watershed resource distribution data, and so on.

[0028] Furthermore, before obtaining the watershed business knowledge document and performing dimension reduction compression encoding on the watershed business knowledge document to obtain the control value function, a unique document identifier is assigned to each watershed business knowledge document, and the document identifier is an integer not greater than the total number of documents.

[0029] Furthermore, the documents in the existing dataset are serially numbered in ascending order starting from 1 to form a combination of [document serial number (or document identifier), document], recorded as [No, FILE_m], where No represents the document identifier, FILE_m represents the document, and 1<=m<=the total number of documents.

[0030] Step S102 , performing empirical mode decomposition on the control value function to obtain a plurality of intrinsic mode function components, and integrating the plurality of intrinsic mode function components into the same coordinate system.

[0031] Step S103 , clustering the watershed business knowledge documents corresponding to the same intrinsic mode function component to obtain a plurality of cluster documents.

[0032] Step S104, obtaining a query document, calculating the Mahalanobis distances between the query point corresponding to the query document and a plurality of intrinsic mode function components, and selecting clustered documents as approximate documents based on the Mahalanobis distances.

[0033] Specifically, the query document uploaded by the terminal device is obtained, and the query document is compressed and encoded by dimensionality reduction to obtain the query point, and then the Mahalanobis distance between the query point and several intrinsic modal function components is calculated, and the approximate document is selected based on the Mahalanobis distance, and the approximate document is fed back to the terminal device as the approximate search result of the query document.

[0034] The present embodiment provides a method for approximate search of watershed business documents. By performing dimensionality reduction compression encoding on watershed business knowledge documents to obtain a comparison value function, the comparison value function is clustered by empirical mode decomposition, and similar documents are clustered, which reduces the number of calculations of document similarity. The Mahalanobis distance between a query point corresponding to a query document and several intrinsic mode function components is calculated, and clustered documents are selected as approximate documents based on the Mahalanobis distance. Only the clustered documents corresponding to the shortest Mahalanobis distance are obtained by comparison, which speeds up the data processing process, solves the problems of slow response speed and low search efficiency of services such as approximate search of documents for watershed business knowledge sharing and search of associated information documents, and significantly improves the document search efficiency and the response efficiency of watershed knowledge sharing.

[0035] In this embodiment, a method for approximate search of flow domain business documents is provided, which can be used in the above-mentioned electronic device. Figure 2 is a flow chart of a method for approximate search of a flow domain business document according to an embodiment of the present invention. Figure 2 As shown, the process includes the following steps: Step S201, obtain a watershed business knowledge document, perform dimensionality reduction compression encoding on the watershed business knowledge document, and obtain a control value function.

[0036] Specifically, the above step S201 includes: Step S2011, using a keyword extraction algorithm to count the target keywords and the frequency of occurrence of the target keywords in each watershed business knowledge document.

[0037] Specifically, the keyword extraction algorithm can use TF-IDF (Term Frequency-Inverse Document Frequency, a common weighted technology used in information retrieval and text mining). Through TF-IDF, the keywords of each basin business knowledge document and the corresponding frequency of each keyword are counted, and the top N (N is an integer) keywords k are selected according to the frequency of occurrence. 1 , k 2 , ..., k N as the target keyword, and the corresponding frequency of each target keyword f 1 、f 2 , ..., f N .

[0038] Step S2012: weighted coding is performed on the target keyword based on the occurrence frequency of the target keyword to obtain a Huffman code.

[0039] Specifically, using the frequency of occurrence of the target keyword f j The target keyword k j Assign weights and obtain the weighted keywords corresponding to the watershed business knowledge document, where j represents the jth target keyword of the watershed business knowledge document.

[0040] Furthermore, Huffman coding is used to encode each weighted keyword according to the occurrence frequency of each target keyword, and the Huffman coding corresponding to each weighted keyword is obtained, that is, the occurrence frequency f corresponding to each target keyword j The corresponding binary code C j .

[0041] Furthermore, the encoding steps of Huffman coding include: counting the frequency of characters: counting the frequency of each character in the weighted keyword; constructing a Huffman tree: constructing a Huffman tree according to the frequency of the characters. The Huffman tree is a special binary tree in which each leaf node represents a character, and on the path from the root node to the leaf node, the left branch is marked as 0 and the right branch is marked as 1; allocating codes: starting from the root node of the Huffman tree, traversing downward along the path until reaching the leaf node, the combination of 0 and 1 on the path is the Huffman code of the corresponding character. Usually, characters with higher probabilities will get shorter codes, while characters with lower probabilities will get longer codes; generating a coding table: recording each character and its corresponding Huffman code in a coding table, which is very important during decoding because the coding table contains key information on how to restore the compressed data to the original data; encoding data: using the generated coding table, each character in the original data is replaced with its corresponding Huffman code, thereby generating compressed data, namely the Huffman code.

[0042] Step S2013, generating a comparison value function based on the target keyword, the occurrence frequency corresponding to the target keyword and the Huffman coding.

[0043] In some optional implementations, the above step S2013 includes: Step a1, based on the target keyword, the occurrence frequency of the target keyword and Huffman coding, the document comparison value corresponding to each watershed business knowledge document is calculated respectively.

[0044] Specifically, the calculation formula of the document comparison value is as follows: (1) in, is the document reference value, For target keywords, For the general Convert to ASIC, is the frequency of occurrence of the target keyword. Huffman coding.

[0045] Step a2, integrating the document comparison values ​​to obtain a comparison value function.

[0046] Specifically, after calculating the document comparison value A corresponding to all basin business knowledge documents j Finally, the timestamps (i.e., document identifiers) of the basin business knowledge documents are integrated to obtain the document comparison value set S = [N o ,A j ], so the set S can be expressed as the following control value function: A=f(N o )(2) Among them, A is the document comparison value function, N o is the document identifier of the j-th document.

[0047] Step S202: Perform empirical mode decomposition on the control value function to obtain a number of intrinsic mode function components, and integrate the several intrinsic mode function components into the same coordinate system. Figure 1 Step S102 of the illustrated embodiment will not be described in detail here.

[0048] Step S203: cluster the watershed business knowledge documents corresponding to the same intrinsic mode function component to obtain a number of cluster documents. Figure 1 Step S103 of the illustrated embodiment will not be described in detail here.

[0049] Step S204, obtain the query document, calculate the Mahalanobis distance between the query point corresponding to the query document and several intrinsic mode function components, and select clustered documents as approximate documents based on the Mahalanobis distance. Figure 1 Step S104 of the illustrated embodiment will not be described in detail here.

[0050] The present embodiment provides an approximate search method for watershed business documents, which utilizes a keyword extraction algorithm to respectively count the target keywords and the occurrence frequency of the target keywords of each watershed business knowledge document, and weights the target keywords according to the occurrence frequency of the target keywords to obtain Huffman coding, thereby realizing dimensionality reduction compression coding of the watershed business knowledge documents, mapping the watershed business knowledge documents into binary coding, and then generating a control value function based on the target keywords, the occurrence frequency corresponding to the target keywords and the Huffman coding, thereby laying a foundation for subsequent empirical mode decomposition clustering.

[0051] In this embodiment, a method for approximate search of flow domain business documents is provided, which can be used in the above-mentioned electronic device. Figure 3 is a flow chart of a method for approximate search of a flow domain business document according to an embodiment of the present invention. Figure 3 As shown, the process includes the following steps: Step S301, obtain the watershed business knowledge document, perform dimension reduction compression coding on the watershed business knowledge document, and obtain the comparison value function. Figure 2 Step S201 of the illustrated embodiment will not be described in detail here.

[0052] Step S302 , performing empirical mode decomposition on the control value function to obtain a plurality of intrinsic mode function components, and integrating the plurality of intrinsic mode function components into the same coordinate system.

[0053] Specifically, the above step S302 includes: Step S3021 , performing empirical mode decomposition on the control value function to obtain the sum of a plurality of intrinsic mode function components and residual terms, wherein the intrinsic mode function components include an upper envelope and a lower envelope.

[0054] Specifically, for the control value function, empirical mode decomposition is used to split it into n IMF (Intrinsic Mode Function) components and a residual term, and each IMF component is regarded as a data class.

[0055] Furthermore, the steps of empirical mode decomposition include: (1) Initialization: Set the original signal X(t), let r(t)=X(t), where r(t) represents the residual signal; initialize the number of iterations i=0 and the order k=1; set the termination threshold condition SD<δ, usually δ is between 0.2 and 0.3.

[0056] (2) Extreme point detection: Find all the maximum and minimum points of the original signal r(t).

[0057] (3) Envelope fitting: The cubic spline interpolation method is used to fit the maximum and minimum points respectively to obtain the upper envelope Uenvelop(t) and the lower envelope Lenvelop(t).

[0058] (4) Calculate the mean envelope: Calculate the mean envelope m(t) of the upper envelope and the lower envelope. The calculation formula is as follows: m(t)=(Uenvelop(t)+Lenvelop(t)) / 2(3) (5) Extracting intermediate signals: Subtracting the mean envelope from the original signal yields the intermediate signal d(t): d(t)=r(t)-m(t)(4) (6) Determine whether IMF conditions are met: Check whether the intermediate signal d(t) satisfies the two conditions of IMF: within the entire data range, the number of local extreme points and zero crossings must be equal or differ by at most one; at any time, the average value of the envelope of the local maximum value and the envelope of the local minimum value is zero.

[0059] If the condition is met, d(t) is an IMF component; if not, repeat steps (2) to (6) until the condition is met.

[0060] (7) Update the remaining signal: If d(t) is an IMF component, denoted as ci(t), and the residual signal is updated, the update formula is as follows: r(t)=r(t)-ci(t)(5) (8) Iterative decomposition: Repeat steps (2) to (7) until r(t) becomes a monotonic function or a constant, or the preset number of iterations is reached.

[0061] (9) Output results: Finally, the original signal X(t) can be expressed as the sum of multiple IMF components and a residual term: X(t)=∑ci(t)+rn(t)(6) Among them, ci(t) is the i-th IMF component and rn(t) is the final residual term.

[0062] Step S3022: for each intrinsic mode function component, respectively calculate the mean of the upper envelope and the lower envelope corresponding to the intrinsic mode function component.

[0063] Step S3023: Use the mean value to replace the horizontal axis of the intrinsic mode function component, and integrate the plurality of intrinsic mode function components into the coordinate system.

[0064] Specifically, the horizontal axis of the intrinsic mode function component is replaced by the mean of the upper envelope and the lower envelope, all IMF components are represented in the same coordinate system, and all documents corresponding to the same intrinsic mode function component are saved in one folder.

[0065] Step S303: cluster the watershed business knowledge documents corresponding to the same intrinsic mode function component to obtain a number of cluster documents. Figure 2 Step S203 of the illustrated embodiment will not be described in detail here.

[0066] Step S304 , obtaining a query document, calculating the Mahalanobis distances between the query point corresponding to the query document and a plurality of intrinsic mode function components, and selecting clustered documents as approximate documents based on the Mahalanobis distances.

[0067] Specifically, the above step S304 includes: Step S3041 , respectively calculating the Mahalanobis distances between the query point corresponding to the query document and a plurality of intrinsic mode function components.

[0068] Specifically, half of the total number of documents is used as the query document identifier of the query document, that is, assuming that the total number of documents is N o , then query document identifier N o _in can be set to , and then according to the above steps S2011 to S2013, the query control value A_in corresponding to the query document is obtained, and the query point can be expressed as [No_in, A_in].

[0069] Furthermore, the query point [No_in, A_in] is represented by a two-dimensional vector as , the intrinsic mode function components are expressed as , and then the mean of the IMF component can be expressed as , where E represents the mathematical expectation calculation, and the calculation formula of the covariance matrix is ​​as follows: (7) in, is the covariance matrix corresponding to the i-th intrinsic mode function component, represents variance calculation, is the document identifier corresponding to the i-th intrinsic mode function component, is the control value corresponding to the i-th intrinsic mode function component.

[0070] Furthermore, the calculation formula of the Mahalanobis distance between the query point and the i-th intrinsic mode function component is as follows: (8) in, is the Mahalanobis distance between the query point and the i-th intrinsic mode function component, is the two-dimensional vector representation of the query point, is the component mean of the i-th intrinsic mode function component, T represents the transpose of the vector, is the covariance matrix corresponding to the i-th intrinsic mode function component.

[0071] Step S3042: taking the intrinsic modal function component corresponding to the minimum Mahalanobis distance as the target intrinsic modal function component, and taking the clustered documents corresponding to the target intrinsic modal function component as the approximate documents.

[0072] Specifically, the IMF component with the smallest Mahalanobis distance is selected, and its corresponding cluster document is returned as the approximate search result.

[0073] The present embodiment provides an approximate search method for watershed business documents. By performing empirical mode decomposition on the control value function, the document clustering problem is converted into a signal decomposition problem, so that the characteristics of the signal at different time scales can be clearly identified and analyzed. Then, several intrinsic mode function components are integrated into the same coordinate system, which can provide a unified framework for subsequent similar document clustering and improve the document search efficiency. Secondly, by calculating the Mahalanobis distance between the query point and several intrinsic mode function components, the document similarity problem is converted into a distance problem between a point and a distribution, which can eliminate the influence of the correlation between variables on the distance calculation, accelerate the data processing process, realize efficient approximate search for similar documents, and reduce the number of document similarity calculations.

[0074] The specific steps of the method for approximate search of watershed business documents are described below through a specific embodiment.

[0075] Embodiment 1: A user browses an article about integrated watershed management in the watershed business management system and wants to learn more about the forms, sources, and remediation methods of watershed pollution. The user initiates a similar document search operation. The specific steps of the similar document search operation include: On the search page, upload the basin comprehensive management document as the query basis. The search system receives the query document uploaded by the user and uses half of the existing document number as its document identifier No_in; Use keyword extraction algorithm to count the target query keywords of the query document and the corresponding occurrence frequency of the target query keywords to obtain weighted query keywords; Encode the weighted query keywords according to the occurrence frequency corresponding to the target query keywords to obtain the Huffman query code corresponding to the weighted query keywords; Calculate the query comparison value of the query document according to the target query keyword, the occurrence frequency corresponding to the target query keyword and the Huffman query code; Obtain the query point [No_in, A_in] corresponding to the query document according to the query document identifier and the query control value of the query document, where A_in is the control value of the query document; The Mahalanobis distance between each intrinsic modal function component and the query document is calculated respectively, and the intrinsic modal function component with the smallest Mahalanobis distance to the query document is selected, and the documents in the folder corresponding to the component are returned and displayed in random order on the page.

[0076] In this embodiment, a device for approximate search of flow domain business documents is also provided, which is used to implement the above-mentioned embodiments and preferred implementation modes, and the descriptions that have been made will not be repeated. As used below, the term "module" can implement a combination of software and / or hardware for a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0077] This embodiment provides a device for approximate search of flow domain business documents, such as Figure 4 As shown, including: Dimensionality reduction compression coding module 401, used to obtain watershed business knowledge documents, perform dimension reduction compression coding on the watershed business knowledge documents, and obtain a comparison value function; An empirical mode decomposition module 402 is used to perform empirical mode decomposition on the control value function to obtain a plurality of intrinsic mode function components, and the plurality of intrinsic mode function components are integrated in the same coordinate system; The clustering module 403 is used to cluster the watershed business knowledge documents corresponding to the same intrinsic mode function component to obtain a plurality of cluster documents; The calculation module 404 is used to obtain a query document, calculate the Mahalanobis distances between the query point corresponding to the query document and a plurality of intrinsic mode function components, and select clustered documents as approximate documents based on the Mahalanobis distances.

[0078] In some optional implementations, the dimension reduction compression coding module 401 includes: A statistical unit, used to respectively count the target keywords and the occurrence frequency of the target keywords in each watershed business knowledge document by using a keyword extraction algorithm; A weighted coding unit, used for weighted coding the target keyword based on the occurrence frequency of the target keyword to obtain a Huffman code; The generating unit is used to generate a comparison value function based on a target keyword, an occurrence frequency corresponding to the target keyword and a Huffman code.

[0079] In some optional implementations, the generating unit includes: A calculation subunit, used to calculate the document comparison value corresponding to each watershed business knowledge document based on the target keyword, the occurrence frequency corresponding to the target keyword and the Huffman coding; The integration subunit is used to integrate the document comparison values ​​to obtain the comparison value function.

[0080] In some optional embodiments, it also includes: The allocation module is used to allocate a unique document identifier to each watershed business knowledge document, and the document identifier is an integer not greater than the total number of documents.

[0081] In some optional implementations, the empirical mode decomposition module 402 includes: A decomposition unit is used to perform empirical mode decomposition on the control value function to obtain the sum of several intrinsic mode function components and residual terms, wherein the intrinsic mode function components include an upper envelope and a lower envelope; A first calculation unit is used to calculate, for each intrinsic mode function component, the mean of the upper envelope and the lower envelope corresponding to the intrinsic mode function component; The integration unit is used to replace the horizontal axis of the intrinsic mode function component with a mean value and integrate a plurality of intrinsic mode function components in a coordinate system.

[0082] In some optional implementations, the calculation module 404 includes: A second calculation unit is used to calculate the Mahalanobis distances between the query point corresponding to the query document and a plurality of intrinsic modal function components respectively; The determination unit is used to use the intrinsic modal function component corresponding to the minimum Mahalanobis distance as the target intrinsic modal function component, and use the clustered documents corresponding to the target intrinsic modal function component as the approximate documents.

[0083] The further functional description of each of the above modules and units is the same as that of the above corresponding embodiments and will not be repeated here.

[0084] In this embodiment, a device for approximate search of watershed business documents is presented in the form of a functional unit, where the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that executes one or more software or fixed programs, and / or other devices that can provide the above functions.

[0085] The embodiment of the present invention also provides a computer device having the above Figure 4 A device for approximate search of watershed business documents is shown.

[0086] See also Figure 5 , Figure 5 is a schematic diagram of the structure of a computer device provided by an optional embodiment of the present invention, such as Figure 5As shown, the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. Various components are connected to each other using different buses for communication, and can be installed on a common mainboard or installed in other ways as needed. The processor can process instructions executed in the computer device, including instructions stored in or on the memory to display the graphical information of the GUI on an external input / output device (such as a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories. Similarly, multiple computer devices can be connected, and each device provides some necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Figure 5 A processor 10 is taken as an example.

[0087] The processor 10 may be a central processing unit, a network processor or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be a dedicated integrated circuit, a programmable logic device or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic or any combination thereof.

[0088] The memory 20 stores instructions executable by at least one processor 10, so that the at least one processor 10 executes the method shown in the above embodiment.

[0089] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to the use of the computer device, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely arranged relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0090] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid state drive; the memory 20 may also include a combination of the above types of memory.

[0091] The computer device also includes an input device 30 and an output device 40. The processor 10, the memory 20, the input device 30 and the output device 40 may be connected via a bus or other means. Figure 5 The example of connecting through bus is taken in the following.

[0092] The input device 30 can receive input digital or character information, and generate key signal input related to the user settings and function control of the computer device, such as a touch screen, a keypad, a mouse, a track pad, a touch pad, an indicator bar, one or more mouse buttons, a trackball, a joystick, etc. The output device 40 may include a display device, an auxiliary lighting device (e.g., an LED) and a tactile feedback device (e.g., a vibration motor), etc. The above-mentioned display device includes but is not limited to a liquid crystal display, a light emitting diode, a display and a plasma display. In some optional embodiments, the display device may be a touch screen.

[0093] The embodiment of the present invention also provides a computer-readable storage medium. The method according to the embodiment of the present invention can be implemented in hardware, firmware, or can be implemented as a computer code that can be recorded in a storage medium, or can be implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and will be stored in a local storage medium through a network download, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state hard disk, etc.; further, the storage medium can also include a combination of the above types of memories. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor, or hardware, the method shown in the above embodiment is implemented.

[0094] A part of the present invention may be applied as a computer program product, such as a computer program instruction, which, when executed by a computer, can call or provide the method and / or technical solution according to the present invention through the operation of the computer. Those skilled in the art should understand that the existence of the computer program instruction in a computer-readable medium includes, but is not limited to, a source file, an executable file, an installation package file, etc., and accordingly, the way in which the computer program instruction is executed by the computer includes, but is not limited to: the computer directly executes the instruction, or the computer compiles the instruction and then executes the corresponding compiled program, or the computer reads and executes the instruction, or the computer reads and installs the instruction and then executes the corresponding installed program. Here, the computer-readable medium may be any available computer-readable storage medium or communication medium accessible to the computer.

[0095] Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A method for approximate search of watershed business documents, characterized in that: The method comprises: Acquire a watershed business knowledge document, perform dimension reduction compression coding on the watershed business knowledge document, and obtain a control value function; Performing empirical mode decomposition on the control value function to obtain a plurality of intrinsic mode function components, wherein the plurality of intrinsic mode function components are integrated in the same coordinate system; Clustering the watershed business knowledge documents corresponding to the same intrinsic mode function component respectively to obtain a plurality of cluster documents; A query document is obtained, Mahalanobis distances between a query point corresponding to the query document and the plurality of intrinsic modal function components are respectively calculated, and the cluster document is selected as an approximate document based on the Mahalanobis distance.

2. The method according to claim 1, characterized in that The process of performing dimension reduction compression coding on the watershed business knowledge document to obtain a control value function includes: Using a keyword extraction algorithm to count the target keywords of each watershed business knowledge document and the frequency of occurrence of the target keywords; Performing weighted coding on the target keyword based on the occurrence frequency of the target keyword to obtain a Huffman code; The control value function is generated based on the target keyword, the occurrence frequency corresponding to the target keyword and the Huffman code.

3. The method according to claim 2, characterized in that The generating the control value function based on the target keyword, the occurrence frequency corresponding to the target keyword and the Huffman coding includes: Calculate the document comparison value corresponding to each watershed business knowledge document based on the target keyword, the occurrence frequency corresponding to the target keyword and the Huffman coding; The document comparison values ​​are integrated to obtain the comparison value function.

4. The method according to claim 1, characterized in that: Before obtaining the watershed business knowledge document and performing dimension reduction compression coding on the watershed business knowledge document to obtain the comparison value function, the method further includes: A unique document identifier is allocated to each of the watershed business knowledge documents, and the document identifier is an integer not greater than the total number of documents.

5. The method according to claim 4, characterized in that The control value function is subjected to empirical mode decomposition to obtain a plurality of intrinsic mode function components, including: Performing empirical mode decomposition on the control value function to obtain the sum of the plurality of intrinsic mode function components and residual terms, wherein the intrinsic mode function components include an upper envelope and a lower envelope; For each of the intrinsic mode function components, respectively calculating the mean of the upper envelope and the lower envelope corresponding to the intrinsic mode function component; The horizontal axis of the intrinsic mode function components is replaced by the mean value, and the plurality of intrinsic mode function components are integrated into the coordinate system.

6. The method according to claim 1, characterized in that The respectively calculating the Mahalanobis distances between the query points corresponding to the query documents and the plurality of intrinsic modal function components, and selecting the clustered documents as approximate documents based on the Mahalanobis distances, comprises: respectively calculating the Mahalanobis distances between the query point corresponding to the query document and the plurality of intrinsic mode function components; The intrinsic modal function component corresponding to the minimum Mahalanobis distance is used as the target intrinsic modal function component, and the clustered document corresponding to the target intrinsic modal function component is used as the approximate document.

7. A device for approximate search of watershed business documents, characterized in that: The device comprises: A dimension reduction compression coding module is used to obtain a watershed business knowledge document, perform dimension reduction compression coding on the watershed business knowledge document, and obtain a comparison value function; An empirical mode decomposition module is used to perform empirical mode decomposition on the control value function to obtain a plurality of intrinsic mode function components, wherein the plurality of intrinsic mode function components are integrated in the same coordinate system; A clustering module, used for clustering the watershed business knowledge documents corresponding to the same intrinsic mode function component to obtain a plurality of cluster documents; The calculation module is used to obtain a query document, calculate the Mahalanobis distances between the query point corresponding to the query document and the plurality of intrinsic mode function components, and select the cluster document as an approximate document based on the Mahalanobis distances.

8. A computer device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the method for approximate search of watershed business documents according to any one of claims 1 to 6 by executing the computer instructions.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the approximate search method for watershed business documents according to any one of claims 1 to 6.

10. A computer program product, characterized in that It comprises computer instructions, and the computer instructions are used to enable a computer to execute the watershed business document approximate search method described in any one of claims 1 to 6.

Images