CN114817378A - Tax file search engine-based retrieval method - Google Patents

Abstract

The invention discloses a retrieval method based on a tax file search engine, when a user enters a system, the corresponding applicable rule and level of a positioning position is determined according to the positioning position of the user, the tax file applicable to the rule and level of the positioning position is displayed in a latest file list, the tax files of central level, province/autonomous region/direct district city level, city/region/state level are displayed on the same screen, and the three rule and level are switchable as TAB columns, and the search engine module is not required to be started repeatedly by combining response type interaction and split screen display, so that the repeated skipping and inefficient retrieval of an operation interface are avoided, the high efficiency of file retrieval and application is realized, and the user is helped to find needed files in the shortest time. The invention supports multiple searches of rule level, time range and keywords singly or in combination.

Description

A Retrieval Method Based on Tax Law Document Search Engine

technical field

The invention relates to a tax law document search engine, in particular to a retrieval method based on the tax law document search engine.

Background technique

There are many types and complex versions of tax law documents. How to find the target document quickly and accurately is a difficult problem that has always plagued tax law practitioners. In this context, the tax law document search engine came into being. However, the design of the existing product interface presentation method, retrieval technology and retrieval method is flawed, resulting in slow retrieval speed or multiple retrievals to obtain retrieval results, which cannot efficiently meet the needs of users to apply tax laws. needs.

Specifically, the shortcomings and deficiencies of the prior art:

1. Restricted by interactive technology, existing products cannot accurately display central-level, provincial/autonomous-municipal-level, city/prefecture/state-level tax law documents on the same screen, and cannot achieve simultaneous retrieval and retrieval of tax law documents at these three effective levels. The results are presented on the same screen; however, users need to accurately apply the tax law documents of these three levels of effectiveness when handling tax matters correctly. Therefore, they have to retrieve, store, and compare multiple times, which is inefficient.

2. Subject to the screen display technology, users of existing products need to compare the old and new regulations or apply them in combination with related regulations. They cannot visually check the old and new regulations or related regulations on a computer screen at the same time. They need to repeatedly minimize the window and do not. Stop switching, the user experience is extremely poor.

3. Existing products directly fetch data from the database and perform calculations each time they are retrieved, which requires high memory usage, high repeated calculations, unsmooth retrieval, slow result feedback, and high concurrency risks, which not only affects user experience, but also causes system risks.

4. Existing product retrieval tax law documents, either rely on self-established retrieval thesaurus, or rely on natural word matching retrieval, retrieval accuracy is not high, retrieval technology is incomplete.

To sum up, the existing products increase the user's time cost and waste. When the user applies the tax law to deal with specific problems in specific work and study, it takes a long time to retrieve documents, the efficiency is not high, the retrieval function is poor, and the user experience is not good.

SUMMARY OF THE INVENTION

The purpose of the present invention is to overcome the deficiencies of the prior art, and to provide a retrieval method based on a tax law document search engine that is more efficient, more stable, and better user experience.

The purpose of this invention is to realize through the following technical solutions:

A retrieval method based on a tax law document search engine, the system defines the regulatory level to which each tax law document belongs when entering tax law documents, and the regulatory levels include central level, province/autonomous region/municipal level, city/region/ State level; when the user enters the system, the applicable regulatory level corresponding to the positioning location is determined according to the user's positioning location, and the tax law documents applicable to the regulatory level for the positioning location are displayed in the latest document list, and the central level, province/autonomous region/regional level is displayed on the same screen Municipal-level, city/prefecture/state-level tax law documents, and the three regulatory levels can be switched as TAB columns. After the switch is selected, the tax law documents of the corresponding regulatory level will be displayed in the file list. ;

Using vue responsive interactive technology, central-level, province/autonomous-region/municipal-level, city/region/state-level documents are presented on the same screen. Municipal-level documents and city/prefecture/state-level documents are screened according to the retrieval conditions, respectively, and the retrieval results after screening of documents at all levels are obtained and presented on the same screen.

The method further includes a step of displaying the contents of tax law documents on a split screen, including the following sub-steps:

S101: When the first tax law file link is opened, a split screen area is established on the basis of the file list, the content of the tax law file is displayed in the split screen area, and the split screen display is started as the first split screen;

S102: When opening the second tax law file link, first determine whether the tax law file has been opened. If so, it will prompt that the file has been opened in the first split screen. If not, further split screen display will be performed on the split screen area. , take the content of the second tax law document as the second split screen, and display it side by side with the first split screen in the split screen area;

S103: When opening the link of the third tax law file, first determine whether the tax law file has been opened. If so, it will prompt the number of screens that the file has been opened. If not, the content of the third tax law file will be taken as The third split screen is displayed in the split screen area, while the first split screen is hidden by sliding, the second split screen and the third split screen are displayed in the split screen area, and a scroll bar for dragging and displaying is generated in the split screen area;

S104: When opening the fourth tax law file link, first determine whether the tax law file has been opened. If it has been opened, it will prompt which sub-screen the file has been opened, and at the same time continue to determine whether the sub-screen is hidden by sliding If it is hidden, it will jump to display the split screen in the split screen area; if it is not opened, the content of the fourth tax law document will be displayed in the split screen area as the fourth split screen, and the second split screen will be displayed at the same time. It is hidden by sliding, and the third and fourth split screens are displayed in the split screen area;

S105: By analogy, until the number of files opened by the user exceeds the threshold of the maximum number of split screens preset by the system, the system prompts that the number of split screens has reached the upper limit.

The split-screen display module realizes the split-screen presentation of a computer screen, and the split-screen technology runs through all modules of the entire system. Using the (Vue swiper sliding page) technology, according to the reading habits of the human eye and the actual needs of work, a computer screen display area can be displayed in three screens at most, and the content of more than three screens can be displayed by dragging the scroll bar; set the screen to hide and Close button to achieve free switching between the first, second and third screens.

The method further includes a preloading and caching step, which associates and caches each tax law document with the file name, document number, full-text keywords, writing time, attachments, regulatory levels, relevant links, etc. of the document. When entering the system for retrieval, if the retrieval conditions match the association of the cache, the result set will be returned directly from the preloaded cache, without the need to fetch data from the database for calculation, and the retrieval results can be quickly fed back to the user. The retrieval is smoother and more efficient, and the system runs more stably.

The method further includes a dynamic caching step, which automatically caches the hot data of the retrieval results under different combinations of retrieval conditions, such as the regulatory level, the time of writing, and the keywords used by the user, so that the different combinations of retrieval conditions are associated with the retrieval results, When the combination of retrieval conditions used by other users matches the combination of retrieval conditions in the cache, the retrieval results are preferentially fed back from the cached hot data, without repeated calculations, the retrieval is smoother and more efficient, and the system runs more stably.

The method also includes a step of automatically updating file retrieval, including the following sub-steps:

S201: Collect tax law documents, and establish a multi-level and multi-dimensional keyword database for tax law documents;

S202: Using the technology (stammering associative participle, which can automatically create vocabulary according to the context of the phrase, to ensure that it is consistent with the actual meaning) technology, cut all tax law documents into words to form the first all thesaurus;

S203: Combine the keyword library and the first all thesaurus to obtain the second all thesaurus;

S204: Using the (stammering custom thesaurus) technology, the keyword database is set as the priority position in the second-cut thesaurus, and the second-cut thesaurus is divided into a keyword database and a non-keyword database;

S205: Using the (Bloom filter, quickly distinguishing keywords and non-keywords) technology to form a two-round combination mode of user retrieval, the user only needs to input a retrieval sentence once, and the retrieval module automatically processes two rounds of retrieval. The processing process is as follows:

Using the (Golang language multi-coroutine concurrent retrieval) technology, the retrieval module uses the keyword database and the non-keyword database to segment the retrieval sentence at the same time. out a number of search terms;

In the first round of retrieval, the (sequence combination) technology is used to match the file name and text content of the tax law documents with different combinations of the main search terms, and the retrieval results are presented in order of matching degree;

If the first round of retrieval cannot meet the user’s needs, the user can activate the “Continue Retrieval” button on the operation interface. The retrieval module will upgrade the search term from the search term to the main search term only in this search, and re-search according to the first round of search rules. And present the retrieval results; if the second round of retrieval can meet the user's needs, the retrieval module automatically records the secondary search terms used in the second round of retrieval.

S206: Use (intelligent search and identification, analyze the search results according to the importance of the files, calculate the weight of the non-keyword search results, the more search results, the greater the weight value, the automatic identification of non-keyword supplements) technology, regularly When the user search starts the second round of search, the word frequency and word vector in the tax law document from the search words (that is, the non-keywords in the thesaurus) used to meet the needs of the user are calculated, and the intelligent dynamic upgrade is supplemented to the keyword database .

The beneficial effects of the present invention are:

1) The present invention supports the accurate presentation of central-level, provincial/autonomous-municipal-level, city/prefecture/state-level tax law documents on the same screen, and realizes the simultaneous retrieval of tax law documents of these three effective levels and the presentation of retrieval results on the same screen. When dealing with tax matters, it can accurately and efficiently apply the tax law documents of these three levels of effectiveness, without the need to retrieve, store, and compare multiple times, which is convenient for users to improve work efficiency.

2) The present invention supports split-screen display, which makes the process of comparing and reading tax law documents more convenient. When retrieving, consulting, and applying tax laws, users can complete the work in split-screen on a computer screen, avoiding repeated minimization of windows and constant switching. Inefficient working state. Split-screen display details such as sliding hiding, prompting, jumping and other functions have a friendly interface, which is convenient for users to improve work efficiency.

3) The present invention supports preloading technology and dynamic caching technology, avoids directly fetching data from the database for each retrieval calculation, solves the problems of high memory occupation, high repeated calculation, unsmooth retrieval, slow result feedback, and high concurrency risk, effectively improving User experience and system risk prevention.

4) The present invention sets up an intelligently upgraded keyword database and a two-round retrieval mode, which comprehensively improves retrieval accuracy, dynamically adapts to the development and changes of the tax law, and meets user retrieval and application requirements.

Detailed ways

The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative efforts shall fall within the protection scope of the present invention.

The invention provides a retrieval method for a tax law document search engine. The user groups applicable to the method include tax-related services of various enterprises and institutions, accounting practitioners, and intermediary agencies (including tax firms, accounting firms, and law firms). Practitioners, cadres of the taxation system, teachers and students of financial and taxation institutions, etc.

A retrieval method based on a tax law document search engine, the system defines the regulatory level to which each tax law document belongs when entering tax law documents, and the regulatory levels include central level, province/autonomous region/municipal level, city/region/ State level; when the user enters the system, the applicable regulatory level corresponding to the positioning location is determined according to the user's positioning location, and the tax law documents applicable to the regulatory level for the positioning location are displayed in the latest document list, and the central level, province/autonomous region/regional level is displayed on the same screen Municipal-level, city/prefecture/state-level tax law documents, and the three regulatory levels can be switched as TAB columns. After the switch is selected, the tax law documents of the corresponding regulatory level will be displayed in the file list. ;

Using vue responsive interactive technology, central-level, province/autonomous-region/municipal-level, city/region/state-level documents are presented on the same screen. Municipal-level documents and city/prefecture/state-level documents are screened according to the retrieval conditions, respectively, and the retrieval results after screening of documents at all levels are obtained and presented on the same screen.

The method further includes a step of displaying the contents of tax law documents on a split screen, including the following sub-steps:

S101: When the first tax law file link is opened, a split screen area is established on the basis of the file list, the content of the tax law file is displayed in the split screen area, and the split screen display is started as the first split screen;

S102: When opening the second tax law file link, first determine whether the tax law file has been opened. If so, it will prompt that the file has been opened in the first split screen. If not, further split screen display will be performed on the split screen area. , take the content of the second tax law document as the second split screen, and display it side by side with the first split screen in the split screen area;

S103: When opening the link of the third tax law file, first determine whether the tax law file has been opened. If so, it will prompt the number of screens that the file has been opened. If not, the content of the third tax law file will be taken as The third split screen is displayed in the split screen area, while the first split screen is hidden by sliding, the second split screen and the third split screen are displayed in the split screen area, and a scroll bar for dragging and displaying is generated in the split screen area;

S104: When opening the fourth tax law file link, first determine whether the tax law file has been opened. If it has been opened, it will prompt which sub-screen the file has been opened, and at the same time continue to determine whether the sub-screen is hidden by sliding If it is hidden, it will jump to display the split screen in the split screen area; if it is not opened, the content of the fourth tax law document will be displayed in the split screen area as the fourth split screen, and the second split screen will be displayed at the same time. It is hidden by sliding, and the third and fourth split screens are displayed in the split screen area;

S105: By analogy, until the number of files opened by the user exceeds the threshold of the maximum number of split screens preset by the system, the system prompts that the number of split screens has reached the upper limit.

The split-screen display module realizes the split-screen presentation of a computer screen, and the split-screen technology runs through all modules of the entire system. Using the (Vue swiper sliding page) technology, according to the reading habits of the human eye and the actual needs of work, a computer screen display area can be displayed in three screens at most, and the content of more than three screens can be displayed by dragging the scroll bar; set the screen to hide and Close button to achieve free switching between the first, second and third screens.

The method further includes a preloading and caching step, which associates and caches each tax law document with the file name, document number, full-text keywords, writing time, attachments, regulatory levels, relevant links, etc. of the document. When entering the system for retrieval, if the retrieval conditions match the association of the cache, the result set will be returned directly from the preloaded cache, without the need to fetch data from the database for calculation, and the retrieval results will be quickly fed back to the user, the retrieval will be smoother and more efficient, and the system will run more stably.

The method further includes a dynamic caching step, which automatically caches the hot data of the retrieval results under different combinations of retrieval conditions, such as the regulatory level, the time of writing, and the keywords used by the user, so that the different combinations of retrieval conditions are associated with the retrieval results, When the combination of retrieval conditions used by other users matches the combination of retrieval conditions in the cache, the retrieval results are preferentially fed back from the cached hot data, without repeated calculations, the retrieval is smoother and more efficient, and the system runs more stably.

The method also includes a step of automatically updating file retrieval, including the following sub-steps:

S201: Collect tax law documents, and establish a multi-level and multi-dimensional keyword database for tax law documents;

S202: Using the technology (stammering associative participle, which can automatically create vocabulary according to the context of the phrase, to ensure that it is consistent with the actual meaning) technology, cut all tax law documents into words to form the first all thesaurus;

S203: Combine the keyword library and the first all thesaurus to obtain the second all thesaurus;

S204: Using the (stammering custom thesaurus) technology, the keyword database is set as the priority position in the second-cut thesaurus, and the second-cut thesaurus is divided into a keyword database and a non-keyword database;

S205: Using the (Bloom filter, quickly distinguishing keywords and non-keywords) technology to form a two-round combination mode of user retrieval, the user only needs to input a retrieval sentence once, and the retrieval module automatically processes two rounds of retrieval. The processing process is as follows:

Using the (Golang language multi-coroutine concurrent retrieval) technology, the retrieval module uses the keyword database and the non-keyword database to segment the retrieval sentence at the same time. out a number of search terms;

In the first round of retrieval, the (sequence combination) technology is used to match the file name and text content of the tax law documents with different combinations of the main search terms, and the retrieval results are presented in order of matching degree;

If the first round of retrieval cannot meet the user’s needs, the user can activate the “Continue Retrieval” button on the operation interface. The retrieval module will upgrade the search term from the search term to the main search term only in this search, and re-search according to the first round of search rules. And present the retrieval results; if the second round of retrieval can meet the user's needs, the retrieval module automatically records the secondary search terms used in the second round of retrieval.

S206: Use (intelligent search and identification, analyze the search results according to the importance of the files, calculate the weight of the non-keyword search results, the more search results, the greater the weight value, the automatic identification of non-keyword supplements) technology, regularly When the user search starts the second round of search, the word frequency and word vector in the tax law document from the search words (that is, the non-keywords in the thesaurus) used to meet the needs of the user are calculated, and the intelligent dynamic upgrade is supplemented to the keyword database .

Practical case:

User A is located in Chengdu. When dealing with VAT-related matters, he needs to use tax law documents at the central, provincial and Chengdu levels at the same time. The latest document list shows the tax law documents of the three effective levels of "Central, Sichuan and Chengdu" on the same screen; and then further selects the category "VAT", you can get one-click access, and the central level, Sichuan provincial level, and Chengdu city are displayed on the same screen. Tax law documents related to value-added tax in three efficacy levels; A user can switch to select the related tax law documents of the three efficacy levels of central level, Sichuan provincial level, and Chengdu level through the TAB column, combined with split-screen presentation, to carry out tax laws of different efficacy levels File association, comparison, application.

Through the combination of regulatory level, time range and keywords, combined with responsive interaction and split-screen display, it is not necessary to repeatedly start the search engine module, avoid repeated jumps and inefficient retrieval of the operation interface, and achieve high efficiency in file retrieval and application.

The foregoing are only preferred embodiments of the present invention, and it should be understood that the present invention is not limited to the forms disclosed herein, and should not be construed as an exclusion of other embodiments, but may be used in various other combinations, modifications, and environments, and Modifications can be made within the scope of the concepts described herein, from the above teachings or from skill or knowledge in the relevant field. However, modifications and changes made by those skilled in the art do not depart from the spirit and scope of the present invention, and should all fall within the protection scope of the appended claims of the present invention.

Claims (6)

1. A retrieval method based on a tax file search engine is characterized in that when a system inputs tax files, a regulation level to which each tax file belongs is defined, wherein the regulation levels comprise a central level, a province/autonomous region/direct administration city level and a city/region/state level; when a user enters a system, determining a rule level corresponding to a positioning position according to the positioning position of the user, displaying tax files suitable for the rule level of the positioning position in a latest file list, displaying the tax files at a central level, province/autonomous region/direct district city level and city/district/state level on the same screen, and switching three rule levels as TAB (tape TAB) columns.

2. The method of claim 1, wherein: the three rule levels serving as TAB bars can be switched, tax files corresponding to the rule levels are displayed in the selected file list after switching, and in the latest file list, the tax files are arranged in descending order according to the writing time of the tax files;

and (3) realizing the on-screen presentation of the central-level, province/autonomous region/direct administration city-level and city/region/state-level files by using vue response type interactive technology, and synchronously screening the central-level files, province/autonomous region/direct administration city-level files and city/region/state-level files according to the retrieval conditions when a user retrieves by using a certain retrieval condition to obtain the retrieval results of the screened files of all levels and presenting the retrieval results on the on-screen presentation.

3. The method of claim 1, wherein: the method also comprises a step of displaying the contents of the tax document in a split screen manner, and comprises the following substeps:

s101: when a first tax file link is opened, a screen splitting area is established on the basis of a file list, and the content of the tax file is displayed in the screen splitting area and is used as a first split screen to start split screen display;

s102: when a second tax file link is opened, firstly judging whether the tax file is opened or not, if so, prompting that the file is opened in a first split screen, if not, further performing split screen display on a split screen area, and displaying the content of the second tax file in the split screen area in parallel with the first split screen left and right by taking the content of the second tax file as a second split screen;

s103: when a third tax file link is opened, firstly, judging whether the tax file is opened or not, if so, prompting that the file is opened in a few split screens, if not, displaying the content of the third tax file as a third split screen into a split screen area, meanwhile, sliding and hiding the first split screen, displaying a second split screen and the third split screen in the split screen area, and generating a scroll bar for dragging and displaying in the split screen area;

s104: when a fourth tax file link is opened, firstly judging whether the tax file is opened or not, if so, prompting that the file is opened in a few sub-screens, meanwhile, continuously judging whether the sub-screens are in a sliding hidden state or not, and if so, skipping to display the sub-screens in a sub-screen area; if the tax file is not opened, displaying the content of the fourth tax file as a fourth split screen into a split screen area, meanwhile, sliding and hiding the second split screen, and displaying a third split screen and a fourth split screen in the split screen area;

s105: and repeating the steps until the number of the files opened by the user exceeds the maximum screen splitting number threshold preset by the system, and prompting that the screen splitting number reaches the upper limit by the system.

4. The method of claim 1, wherein: and the method also comprises a preloading caching step, wherein each tax file is associated and cached with the file name, the character number, the full-text key words, the literary composition time, the attachment, the regulation level and the related links of the file, and when a user enters a system for retrieval, if the retrieval condition is matched with the cached association, a result set is directly returned from the preloading caching.

5. The method of claim 1, wherein: the method also comprises a dynamic caching step, wherein the retrieval result hot data under different combination retrieval conditions of the rule level, the literary composition time and the key word used by the user is automatically cached, so that different retrieval condition combinations are associated with the retrieval result, and when the retrieval condition combinations used by other users are matched with the cached retrieval condition combinations, the retrieval result is preferentially fed back from the cached hot data.

6. The method of claim 1, wherein: the method also comprises a step of automatically upgrading file retrieval, which comprises the following sub-steps:

s201: collecting tax documents, and establishing a tax document keyword library in multiple layers and multiple dimensions;

s202: cutting words of all tax documents to form a first word stock;

s203: merging the keyword library and the first keyword library to obtain a second keyword library;

s204: setting the keyword library as a priority status in a second word segmentation library, and splitting the second word segmentation library into the keyword library and a non-keyword library;

s205: the method is characterized in that keywords and non-keywords are rapidly distinguished to form a two-wheel combination mode of user retrieval, a user only needs to input a retrieval statement once, the retrieval module automatically processes two-wheel retrieval, and the processing process is as follows:

the retrieval module uses the keyword library and the non-keyword library to cut words of the retrieval sentences simultaneously, cuts out a plurality of main retrieval words according to the keyword library, and cuts out a plurality of auxiliary retrieval words according to the non-keyword library;

during the first round of retrieval, different combinations formed by the main retrieval words are matched with the file names and the text contents of the tax file, and retrieval results are presented in sequence according to the matching degree;

if the first round of retrieval can not meet the requirements of the user, the user continues to retrieve, the retrieval module only upgrades the secondary retrieval words into the main retrieval words in the retrieval, and the retrieval module retrieves again according to the first round of retrieval rules and presents the retrieval results;

if the second round of retrieval can meet the requirements of the user, the retrieval module automatically records the secondary retrieval words used in the second round of retrieval;

s206: and (3) intelligent retrieval identification, analyzing retrieval results according to the importance of the file, calculating the weight of non-keyword retrieval results, automatically identifying non-keyword supplementation when the more the retrieval results are, the larger the obtained weight is, and calculating the word frequency and word vector of the retrieval words in the tax file which meet the requirements of a user when starting a second round of retrieval for the user retrieval at regular intervals, so that the words are intelligently and dynamically upgraded and supplemented into a keyword library.