Enterprise information data retrieval system

Abstract

The invention discloses an enterprise information data retrieval system, which belongs to the technical field of information data retrieval, and comprises a data storage module, a data retrieval module, a permission endowing module, an index guide module and an associated data supplement module, the data storage module is used for respectively establishing enterprise storage databases for storing enterprise related data, the data retrieval module comprises a retrieval algorithm and is used for quickly retrieving the data in the data storage module according to information input by a user, and the authority endowing module. According to the enterprise information data retrieval system, the functions of information retrieval, keyword extraction, data compression, integrated data storage, authority control, index optimization and external data supplementation can be achieved, a comprehensive, efficient and safe information retrieval platform is provided for an enterprise, employees inside the enterprise can quickly retrieve information conveniently, and the enterprise efficiency is improved. And the response speed of information retrieval is improved.

Description

Technical Field

[0001] The invention belongs to the technical field of information data retrieval, and in particular is an enterprise information data retrieval system. Background Art

[0002] Enterprise information data refers to the digital representation of various information related to the enterprise. It covers the data generated, collected and used by the enterprise in various aspects such as operation, management and decision-making, including financial statements, receipts and invoices, product information, employee files, etc. As the business time of an enterprise continues to grow, the relevant internal information data of the enterprise will become redundant and complicated. When enterprise employees search for internal information, there is often a problem of slow response. The existing enterprise information data retrieval system cannot extract and compress the complicated data generated in the process of enterprise operation, which is not conducive to the rapid retrieval of enterprise information data. There are problems of low practicality and functionality. In addition, it is impossible to regularly supplement the enterprise information database with relevant information according to the production content and R&D direction of the enterprise, and it cannot provide convenience for the R&D of the enterprise; In view of the above, this case proposes an enterprise information data retrieval system, which solves the above technical problems by adding an index guidance module and an associated data supplement module. Summary of the invention

[0003] The present invention aims to solve at least one of the technical problems existing in the prior art. To this end, the present invention proposes an enterprise information data retrieval system to solve the above technical problem by improving the detection method and processing method.

[0004] In order to achieve the above object, the present invention adopts the following technical solutions: An enterprise information data retrieval system includes a data storage module, a data retrieval module, a permission granting module, an index guiding module, and a related data supplementing module; The data storage module establishes enterprise storage databases for storing enterprise-related data based on enterprise data types; The data retrieval module includes a retrieval algorithm for quickly retrieving data in the data storage module according to information input by the user; The authority granting module establishes employee accounts and grants corresponding search permissions based on the employee's relevant information, and marks the permission level of the data in the data storage module; The index guidance module sets a time limit threshold, performs keyword extraction and compression on data that exceeds the time threshold, and sets different keyword extraction strategies according to different data types, including receipts, information, and archives, using different extraction strategies for keyword extraction; The associated data supplement module regularly searches for external data based on the enterprise's R&D direction, R&D content, and product content, and supplements the enterprise's internal information storage database.

[0005] Furthermore, the data storage module uses the server resources provided by the cloud computing platform for data storage, and creates different types of databases for data storage according to the use requirements of the enterprise, including relational databases and non-relational databases, wherein: Relational databases are used to store structured data, including employee information and financial records. Establish a relational database and collect employee information and financial records into the database. Employee information includes name, position, age, and work location. Financial records include invoices, bills, reimbursement records, and public transfer records. Non-relational databases are used to store semi-structured or unstructured data, including document data and log data. A non-relational database is established on the cloud service platform to store internal enterprise document data and log data; The data retrieval module, based on the information input by the user, uses Boolean logic operators to Combine multiple search terms input by users to search and match the data in the relevant database in the data storage module.

[0006] Furthermore, the index guidance module sets a time limit threshold, performs keyword extraction and compression on data that exceeds the time limit, and sets different keyword extraction strategies according to different data types, including receipts, information, and archives, and uses different extraction strategies to extract keywords, including the following steps: According to the company's existence and data type, different time thresholds are set to classify the relevant data in the company's storage database; Based on the different types of data in the enterprise storage database, different extraction strategies are adopted for receipt, information and archive data. The algorithm is used for compression to reduce data storage volume and retrieval burden.

[0007] Furthermore, different time thresholds are set according to the existence years and data types of the enterprise, and the relevant data in the enterprise storage database is classified. The specific steps are: Based on the company's years of existence , set the benchmark time threshold ,in: When 0≤ ≤5, the benchmark time threshold ; When 5< ≤10, the benchmark time threshold ; when >10, the benchmark time threshold ; Based on different data types, for the benchmark time threshold Make corrections, including receipt, information, and file data. The specific steps are as follows: For receipt data, the receipt time threshold , for all the existing time in the enterprise storage database greater than The receipt data is marked in blue; For information data, the information time threshold , for all the existing time in the enterprise storage database greater than Information data are marked in blue; For archival data, the archival time threshold , for all the existing time in the enterprise storage database greater than The archival data is marked in blue.

[0008] Furthermore, based on the different types of data in the enterprise storage database, different extraction strategies are adopted for receipt, information and archive data. The algorithm is used to compress data to reduce the storage volume and retrieval burden. The specific steps are as follows: For receipt data, including invoices and receipts, the title of the receipt data is extracted through field recognition, and the amount and date in the receipt data are extracted based on regular expressions. The extracted data is combined in the form of title-date-amount. For information data, including R&D materials and experimental data, the text is broken down into vocabulary units and the TF-IDF value of each word is calculated to identify important keywords. The specific steps are as follows: For each word in the document, calculate the number of times it appears in the document. The algorithm formula is: ;in, represents the frequency of word i in document j, Represents the number of times word i appears in document j, The sum of the number of occurrences of all words in document j; Calculate the inverse document frequency of word i. The algorithm formula is: ;in, Represents the total number of documents, represents the inverse document frequency of word i, Represents the number of documents for word i; Get the TF-IDF value of related words. The algorithm formula is: ;in, The TF-IDF weight of the representative word i in the document j is sorted in descending order, and the top three words are selected as the keywords of the relevant documents; For archival data, the NER model is used to extract the employee name, employee number, and position information associated with the archive as keywords for the relevant archives; pass The algorithm compresses the blue marked data and uses keywords to name the related compressed files.

[0009] Furthermore, the associated data supplement module periodically searches for external data based on the enterprise's R&D direction, R&D content, and product content, and supplements the enterprise's internal information storage database, including the following steps: Collect enterprise-related data content, based on the set time interval t, collect external related information data, including magazines, journals, patents, and supplement the enterprise storage database; Based on the collected external related information and combined with the keywords in the enterprise storage database, the collected information data is matched and associated.

[0010] Furthermore, the collection of enterprise-related data content is based on a set time interval t, and external related information data is collected, including magazines, journals, and patents, to supplement the enterprise storage database. The specific steps are: Collect enterprise-related data content, including enterprise products, R&D direction, R&D content, and product content, extract keywords from enterprise-related data content, and crawl external data based on the set time interval t; Based on the extracted keywords of enterprise-related data content, crawlers are used to capture keyword-related journal articles and magazine content. Academic databases include , patent database includes And add the crawled article data into the enterprise storage database to establish an external document archive.

[0011] Furthermore, the collected external related information is combined with keywords in the enterprise storage database to match and associate the collected information data, and the specific steps are: Based on the relevant data in the external literature database, keywords are extracted for enterprise products, R&D directions, R&D contents, and product contents. By performing TF-IDF calculations on enterprise data and literature databases, keywords are obtained and matched. Relevant external literature information, including numbers and titles, is marked at the relevant data in the relevant enterprise storage database.

[0012] Compared with the prior art, the present invention has the following beneficial effects: In the present invention, by setting an index guidance module, a reference time threshold is obtained based on the enterprise's years of experience, and different characteristic time thresholds are set for different data in combination with the type of internal enterprise data. Keyword extraction and compression are performed on data whose existence time is greater than the threshold, thereby improving the response speed and efficiency of information retrieval. Through keyword extraction and compression, the amount of operations on the original data is reduced, and the calculation overhead is reduced; In the present invention, by setting a time threshold, data with a long time or no importance can be filtered out, thereby reducing the amount of data to be retrieved and improving the retrieval speed. With the help of the index guidance module, a more efficient index structure can be constructed to accelerate the data retrieval process, improve information retrieval efficiency, and enhance the practicality of the system; In the present invention, different keyword extraction strategies are adopted for different data types. Using special keyword extraction strategies according to different types of data can ensure that the extracted keywords are more targeted and effective, thereby improving the data extraction accuracy and enhancing the accuracy of subsequent retrieval. The extracted keywords are compressed, which can effectively reduce the storage space occupied and significantly reduce the data volume without losing important information. In the present invention, by combining the enterprise's R&D direction, R&D content, and product content, external data is regularly searched and supplemented, including magazines, journals, and patents, and keywords are linked to relevant information in the enterprise's internal database, so that the enterprise's R&D personnel can obtain the latest technology trends and research results when performing information retrieval, so that the R&D personnel can use the enterprise's existing resources more efficiently, and the functionality of the system is enhanced; The entire enterprise information data retrieval system can realize information retrieval, keyword extraction, data compression, integrated data storage, permission control, index optimization and external data supplement functions, providing the enterprise with a comprehensive, efficient and secure information retrieval platform, facilitating the rapid retrieval of information by internal employees and improving the response speed of information retrieval. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 This is a block diagram of an enterprise information data retrieval system of the present invention. DETAILED DESCRIPTION

[0014] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0015] Example 1: Figure 1 As shown, an enterprise information data retrieval system includes a data storage module, a data retrieval module, a permission granting module, an index guiding module, and a related data supplementing module; The data storage module includes a cloud server, and based on the enterprise data type, establishes enterprise storage databases for storing enterprise-related data; The data storage module uses the server resources provided by the cloud computing platform for data storage. According to the needs of the enterprise, different types of databases are created for data storage, including relational databases and non-relational databases, including: Relational databases are used to store structured data, including employee information and financial records. Establish a relational database and collect employee information and financial records into the database. Employee information includes name, position, age, and work location. Financial records include invoices, bills, reimbursement records, and public transfer records. Non-relational databases are used to store semi-structured or unstructured data, including document data and log data. A non-relational database is established on the cloud service platform to store internal enterprise document data and log data; It should be noted that the specific database type needs to be established according to the needs of the enterprise. Relational databases support complex queries and transaction management, while non-relational databases are suitable for large-scale data storage and fast read and write operations. It is an open source relational database that can support complex queries. The data retrieval module includes a retrieval algorithm for quickly retrieving the data in the data storage module according to the information input by the user, and based on the Boolean logic operator, Combine multiple search terms input by users to search and match the data in the relevant database in the data storage module.

[0016] The authority granting module creates employee accounts and grants corresponding search permissions based on employee related information. At the same time, the authority level of the data in the data storage module is marked, allowing employee accounts that meet the authority to access data at the corresponding authority level; It should be noted that through the administrator account, employee accounts are marked with permission levels, including S, A, B, and C. At the same time, the data in the data storage module is marked with levels, including S, A, B, and C. Among them, S-level employee accounts can access S, A, B, and C-level data, A-level employee accounts can access A, B, and C-level data, B-level employee accounts can access B and C-level data, and C-level employee accounts can only access C-level data.

[0017] The index guidance module sets a time limit threshold, extracts and compresses keywords for data that exceeds the time threshold, and sets different keyword extraction strategies according to different data types, including receipts, information, and archives. Different extraction strategies are used for keyword extraction, including the following steps: According to the existence years and data types of the enterprise, different time thresholds are set to classify the relevant data in the enterprise storage database. The specific steps are as follows: Based on the company's years of existence , set the benchmark time threshold ,in: When 0≤ ≤5, the benchmark time threshold ; When 5< ≤10, the benchmark time threshold ; when >10, the benchmark time threshold ; Based on different data types, for the benchmark time threshold Make corrections, including receipt, information, and file data. The specific steps are as follows: For receipt data, the receipt time threshold , for all the existing time in the enterprise storage database greater than The receipt data is marked in blue; For information data, the information time threshold , for all the existing time in the enterprise storage database greater than Information data are marked in blue; For archival data, the archival time threshold , for all the existing time in the enterprise storage database greater than The archival data is marked in blue.

[0018] It should be noted that setting different time thresholds for different types of data can ensure that important data is retained intact in the enterprise storage database to the greatest extent possible, and as much non-important data as possible is processed by extracting keywords and compressing them, thereby reducing the response speed and retrieval efficiency of relevant personnel during the retrieval process, and effectively processing and optimizing non-important data, which helps to improve storage efficiency and retrieval performance, so that relevant personnel can obtain the required information more quickly when retrieving data, while saving storage resources.

[0019] Based on the different types of data in the enterprise storage database, different extraction strategies are adopted for receipt, information and archive data. The algorithm is used to compress data to reduce the storage volume and retrieval burden. The specific steps are as follows: For receipt data, including invoices and receipts, the title of the receipt data is extracted through field recognition, and the amount and date in the receipt data are extracted based on regular expressions. The extracted data is combined in the form of title-date-amount. For information data, including R&D materials and experimental data, the text is broken down into vocabulary units and the TF-IDF value of each word is calculated to identify important keywords. The specific steps are as follows: For each word in the document, calculate the number of times it appears in the document. The algorithm formula is: ;in, represents the frequency of word i in document j, Represents the number of times word i appears in document j, The sum of the number of times all words appear in document j. It should be noted that for each word in a document, the number of times it appears in the document is calculated. The more times it appears, the higher the word frequency, which means that the word is more important in the document.

[0020] Calculate the inverse document frequency of word i. The algorithm formula is: ;in, Represents the total number of documents, represents the inverse document frequency of word i, Represents the number of documents for word i; it should be noted that Represents the total number of documents, that is, the number of documents in the document set, Represents the inverse document frequency of word i, which measures the rarity of word i in the entire document set. The number of documents that represent word i, that is, the number of documents that contain word i in the document set D. d represents the document. The inverse document frequency reflects the scarcity of the word in the entire document set. If a word appears in many documents, its IDF value will be lower. Conversely, if a word appears in only a few documents, its IDF value will be higher.

[0021] Get the TF-IDF value of related words. The algorithm formula is: ;in, It represents the TF-IDF weight of word i in document j. The TF-IDF weights of the words in the document are arranged in descending order, and the top three are selected as the keywords of the relevant documents. It should be noted that the higher the TF-IDF weight, the more important the word is in a specific document. That is, if a word appears frequently in a document but rarely in other documents, the TF-IDF value of this word will be very high. A high TF-IDF value indicates that this word contributes greatly to the content of the document.

[0022] For archival data, the NER model is used to extract the employee name, employee number, and position information associated with the archive as keywords for the relevant archives; It should be noted that named entity recognition, or NER model, is an important technology in natural language processing. It can identify specific types of entities from text and classify the entities into predefined categories. NER usually includes entity categories such as names of people, places, names of organizations, dates and times, etc.

[0023] pass The algorithm compresses the blue marked data and uses keywords to name the related compressed files.

[0024] Embodiment 2: A related data supplementation module regularly searches for external data based on the R&D direction, R&D content, and product content of the enterprise, and supplements the enterprise's internal information storage database, including the following steps: Collect enterprise-related data content. Based on the set time interval t, collect external related information data, including magazines, journals, and patents, to supplement the enterprise storage database. The specific steps are: Collect enterprise-related data content, including enterprise products, R&D direction, R&D content, and product content, extract keywords from enterprise-related data content, and crawl external data based on the set time interval t; Based on the extracted keywords of enterprise-related data content, crawlers are used to capture keyword-related journal articles and magazine content. Academic databases include , patent database includes And add the crawled article data into the enterprise storage database to establish an external document archive.

[0025] It should be noted that the time interval t is usually set to 6 months, and can also be flexibly set according to the R&D content of the enterprise. By setting different time intervals t, the speed of acquiring external literature and the speed of supplementing the enterprise database can be adjusted.

[0026] Based on the collected external related information and combined with the keywords in the enterprise storage database, the collected information data is matched and associated. The specific steps are as follows: Based on the relevant data in the external literature database, keywords are extracted for enterprise products, R&D directions, R&D contents, and product contents. By performing TF-IDF calculations on enterprise data and literature databases, keywords are obtained and matched. Relevant external literature information, including numbers and titles, is marked at the relevant data in the relevant enterprise storage database.

[0027] The enterprise information data retrieval system of the present invention, when in use, establishes an employee account and grants corresponding retrieval authority based on the relevant information of the employee, marks the authority level of the data in the data storage database, and quickly retrieves the data in the enterprise storage database according to the information input by the employee through a full-text search algorithm, a rule-based retrieval algorithm and a fuzzy matching algorithm; Based on the company's years of operation and the type of related data in the company's storage database, keyword extraction and compression are performed on data that is greater than the time threshold, providing company employees with a better response rate and retrieval efficiency when performing information retrieval. At the same time, based on the company's R&D content and product content, external related literature, journals, and patent data are regularly crawled, and external document archives are established to match related documents in the company's database to facilitate retrieval by company R&D personnel.

[0028] In the embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of modules is only a logical function division, and there may be other division methods in actual implementation; the modules described as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the method of this embodiment.

[0029] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. An enterprise information data retrieval system, characterized in that: It includes data storage module, data retrieval module, authority granting module, index guidance module and related data supplement module; The data storage module establishes enterprise storage databases for storing enterprise-related data based on enterprise data types; The data retrieval module includes a retrieval algorithm for quickly retrieving data in the data storage module according to information input by the user; The authority granting module establishes employee accounts and grants corresponding search permissions based on the employee's relevant information, and marks the permission level of the data in the data storage module; The index guidance module sets a time limit threshold, performs keyword extraction and compression on data that exceeds the time threshold, and sets different keyword extraction strategies according to different data types, including receipts, information, and archives, using different extraction strategies for keyword extraction; The associated data supplement module regularly searches for external data based on the enterprise's R&D direction, R&D content, and product content, and supplements the enterprise's internal information storage database.

2. The enterprise information data retrieval system according to claim 1, characterized in that: The data storage module uses the server resources provided by the cloud computing platform for data storage. According to the needs of the enterprise, different types of databases are created for data storage, including relational databases and non-relational databases, wherein: Relational databases are used to store structured data, including employee information and financial records. Establish a relational database and collect employee information and financial records into the database. Employee information includes name, position, age, and work location. Financial records include invoices, bills, reimbursement records, and public transfer records. Non-relational databases are used to store semi-structured or unstructured data, including document data and log data. A non-relational database is established on the cloud service platform to store internal enterprise document data and log data; The data retrieval module, based on the information input by the user, uses Boolean logic operators to Combine multiple search terms input by users to search and match the data in the relevant database in the data storage module.

3. The enterprise information data retrieval system according to claim 1, characterized in that: The index guidance module sets a time limit threshold, extracts and compresses keywords for data that exceeds the time limit, and sets different keyword extraction strategies according to different data types, including receipts, information, and archives, and uses different extraction strategies to extract keywords, including the following steps: According to the company's existence and data type, different time thresholds are set to classify the relevant data in the company's storage database; Based on the different types of data in the enterprise storage database, different extraction strategies are adopted for receipt, information and archive data. The algorithm is used for compression to reduce data storage volume and retrieval burden.

4. The enterprise information data retrieval system according to claim 3, characterized in that: According to the existence years and data types of the enterprise, different time thresholds are set respectively to classify the relevant data in the enterprise storage database. The specific steps are as follows: Based on the company's years of existence , set the benchmark time threshold ,in: When 0≤ ≤5, the benchmark time threshold ; When 5< ≤10, the benchmark time threshold ; when >10, the benchmark time threshold ; Based on different data types, for the benchmark time threshold Make corrections, including receipt, information, and file data. The specific steps are as follows: For receipt data, the receipt time threshold , for all the existing time in the enterprise storage database greater than The receipt data is marked in blue; For information data, the information time threshold , for all the existing time in the enterprise storage database greater than Information data are marked in blue; For archival data, the archival time threshold , for all the existing time in the enterprise storage database greater than The archival data is marked in blue.

5. The enterprise information data retrieval system according to claim 3, characterized in that: Based on the different types of data in the enterprise storage database, different extraction strategies are adopted for receipt, information and archive data. The algorithm is used to compress data to reduce the storage volume and retrieval burden. The specific steps are as follows: For receipt data, including invoices and receipts, the title of the receipt data is extracted through field recognition, and the amount and date in the receipt data are extracted based on regular expressions. The extracted data is combined in the form of title-date-amount. For information data, including R&D materials and experimental data, the text is broken down into vocabulary units and the TF-IDF value of each word is calculated to identify important keywords. The specific steps are as follows: For each word in the document, calculate the number of times it appears in the document. The algorithm formula is: ;in, represents the frequency of word i in document j, Represents the number of times word i appears in document j, The sum of the number of occurrences of all words in document j; Calculate the inverse document frequency of word i. The algorithm formula is: ;in, Represents the total number of documents, represents the inverse document frequency of word i, Represents the number of documents for word i; Get the TF-IDF value of related words. The algorithm formula is: ;in, The TF-IDF weight of the representative word i in the document j is sorted in descending order, and the top three words are selected as the keywords of the relevant documents; For archival data, the NER model is used to extract the employee name, employee number, and position information associated with the archive as keywords for the relevant archives; pass The algorithm compresses the blue marked data and uses keywords to name the related compressed files.

6. The enterprise information data retrieval system according to claim 1, characterized in that: The associated data supplement module regularly searches for external data based on the R&D direction, R&D content, and product content of the enterprise, and supplements the enterprise's internal information storage database, including the following steps: Collect enterprise-related data content, based on the set time interval t, collect external related information data, including magazines, journals, patents, and supplement the enterprise storage database; Based on the collected external related information and combined with the keywords in the enterprise storage database, the collected information data is matched and associated.

7. An enterprise information data retrieval system according to claim 6, characterized in that: The collection of enterprise-related data content is based on a set time interval t, and external related information data is collected, including magazines, journals, and patents, to supplement the enterprise storage database. The specific steps are: Collect enterprise-related data content, including enterprise products, R&D direction, R&D content, and product content, extract keywords from enterprise-related data content, and crawl external data based on the set time interval t; Based on the extracted keywords of enterprise-related data content, crawlers are used to capture keyword-related journal articles and magazine content. Academic databases include , patent database includes And add the crawled article data into the enterprise storage database to establish an external document archive.

8. An enterprise information data retrieval system according to claim 7, characterized in that: The collected external related information is combined with keywords in the enterprise storage database to match and associate the collected information data. The specific steps are: Based on the relevant data in the external literature database, keywords are extracted for enterprise products, R&D directions, R&D contents, and product contents. By performing TF-IDF calculations on enterprise data and literature databases, keywords are obtained and matched. Relevant external literature information, including numbers and titles, is marked at the relevant data in the relevant enterprise storage database.

Images