Customer archive digital management method and device based on mining technology, and electronic equipment

Abstract

The invention discloses a customer archive digital management method and device based on a mining technology and electronic equipment, and relates to the field of archive management, and the method comprises the following steps: 1, according to an archive digital target and range, defining an archive type, a field and an identification feature which need to be identified, and completing the construction of an identification framework; step 2, scanning an entity file according to a preset recognition framework, preprocessing an obtained scanning image, and performing OCR character conversion to obtain entity data; automatic processing, including feature extraction, classification and filtering, is performed on client entity and virtual file data by using a data mining technology, so that the time and cost of manual operation are greatly reduced, the speed and efficiency of data processing are improved, one-way fusion of entity features and virtual features is realized, and the efficiency of data processing is improved. Virtual data are effectively integrated into entity archives by using a data mining technology, relevance of archive data of different sources and representation forms is constructed, and deep customer archive information is provided.

Description

Technical Field

[0001] The present invention relates to the technical field of archive management, and in particular to a method, device and electronic equipment for digital management of customer archives based on mining technology. Background Art

[0002] In the field of archive management, with the exponential growth of data volume and the continuous improvement of the information level of company systems, traditional customer management systems are facing huge challenges. The data processing capabilities of these systems are insufficient and it is difficult to meet the growing demand for data storage. In addition, they are unable to cope with data analysis and utilization. The introduction of big data technology provides new ideas and methods to solve the above problems. With its powerful data processing capabilities, efficient data analysis capabilities and flexible data storage mechanisms, big data technology provides solid technical support for the design and implementation of customer archive digital management systems.

[0003] However, due to the complexity of customer data sources and the inconsistency of statistical methods, the actual archives obtained are also inconsistent, divided into physical archives and virtual archives. It is difficult to extract and analyze the features of archive data of different types, especially the existence of physical data. As a result, in the statistical process, it is extremely difficult to quickly merge the physical data of the same customer with its own virtual data in a short period of time, which is not conducive to customer archive analysis and research;

[0004] Therefore, a method, device and electronic device for digital management of customer files based on mining technology are needed to solve the above problems. Summary of the invention

[0005] 1. Technical issues to be resolved

[0006] In view of the above-mentioned shortcomings of the prior art, the present invention provides a customer file digital management method, device and electronic equipment based on mining technology, which can effectively solve the problems of the prior art.

[0007] (II) Technical solution

[0008] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0009] The present invention discloses a method for digital management of customer files based on mining technology, comprising the following steps:

[0010] Step 1: Based on the goal and scope of archive digitization, define the archive types, fields and identification features that need to be identified, and complete the construction of the identification framework;

[0011] Step 2: Scan the physical file according to the preset recognition framework, obtain the scanned image, perform OCR character conversion after pre-processing, and obtain the physical data;

[0012] Step 3: Extract a number of entity features from the entity data through natural language processing technology and group them according to the established classification criteria;

[0013] Step 4: Sort the grouped features by their criticality, and prioritize the highly critical features for indexable item definition;

[0014] Step 5: Obtain several indexable items and search them on the designated searchable platform, match them in the defined index items according to the user's query conditions, and extract relevant virtual archive data;

[0015] Step 6: Preset user demand indicators of time range, data quality and relevance score, and filter virtual archive data based on demand indicators combined with data mining technology;

[0016] Step 7: From the filtered virtual data, extract the virtual features related to the original entity profile and perform one-way fusion with the extracted entity features. The fused features are output as the total features of the customer profile.

[0017] Furthermore, the process of extracting relevant virtual archive data in step 5 is as follows:

[0018] Select relevant items from a set of predefined index items based on the user's query conditions;

[0019] Use defined query conditions and index items to search for data on the selected searchable platform;

[0020] From the search results returned by the platform, filter by analyzing the matching degree with the user's query conditions, the release time of the data and the relevance of the data;

[0021] The screened virtual archive data will be sorted and fed back.

[0022] Furthermore, during the OCR character conversion process in step 2, the scanned image is segmented into recognizable independent units, the information of each independent unit is identified by an image information recognition method, parameter compensation is performed, and the data is compared and analyzed with the data in a preset standard archive information library. The coded content of the information is determined based on the analysis results and stored in the database in a specified format.

[0023] Furthermore, in the parameter compensation process, based on the recognition result of the independent unit information, the acquisition accuracy of the original file is controlled. The specific calculation process is:

[0024]

[0025] In the formula, V1 and V2 represent the two boundary values ​​of the recognition cursor, V orepresents the median value of the identification cursor, k represents the calculation constant, a represents the compensation parameter, S i represents the relative measurement result of archival image recognition, and g represents the standard grayscale of paper archives.

[0026] Furthermore, the data mining process in step 6 is:

[0027] Acquire virtual archive data within a specific time window and verify the quality of virtual archive data through data cleaning techniques and data verification rules;

[0028] Build a machine learning model to analyze keyword matching, content relevance, and user behavior data factors, obtain the degree of matching between the current virtual data and the query purpose, and give priority to highly matching virtual data.

[0029] Furthermore, the attributes of the virtual features in step 7 include: structured data, semantic information or digital content corresponding to the physical archive.

[0030] The customer file digital management device based on mining technology includes:

[0031] The management module, as the core control of the archive management terminal, edits and sends control instructions to each functional module;

[0032] The framework building module is used to identify the types, fields and features of archives that need to be digitized and establish an identification framework based on the goals and scope of archive digitization;

[0033] A scanning conversion module, used for scanning physical files or materials according to the recognition framework, generating scanned images, and converting the images into textual physical data;

[0034] An entity feature extraction module is used to extract entity features from text entity data, and classify and group the entity features;

[0035] The criticality sorting module is used to sort the criticality of the extracted entity features and recommend key features first. The features recommended first are defined as indexable items.

[0036] The virtual retrieval module is used to select relevant items from predefined indexable items according to the user's query conditions, perform data retrieval, filter search results by analyzing the matching degree with the query conditions, the release time of the data and the relevance, and extract the related virtual archive data;

[0037] The demand filtering module is used to filter virtual archive data according to the user's demand indicators, build a machine learning model to analyze the matching degree between virtual data and the query purpose, and give priority to highly matching virtual data;

[0038] One-way fusion module, used to unidirectionally fuse highly matched virtual data with extracted entity features and output the fused total features;

[0039] The storage module is used as a data storage terminal to receive, transmit and back up interactive data, collected data and analyzed data in the cloud.

[0040] Furthermore, the management module is interactively connected to the framework construction module, the scan conversion module and the storage module through a wireless network, the scan conversion module is interactively connected to the entity feature extraction module through a wireless network, the entity feature extraction module is interactively connected to the criticality sorting module through a wireless network, the criticality sorting module is interactively connected to the virtual retrieval module through a wireless network, the virtual retrieval module is interactively connected to the demand filtering module through a wireless network, the demand filtering module is interactively connected to the one-way fusion module through a wireless network, and the one-way fusion module is interactively connected to the storage module through a wireless network.

[0041] An electronic device comprises a processor and a memory, wherein the memory stores at least one instruction, at least one program, a code set or an instruction set, and the at least one instruction, at least one program, the code set or the instruction set is loaded and executed by the processor to implement a customer profile digital management method based on mining technology.

[0042] A computer-readable storage medium stores at least one instruction, at least one program, a code set or an instruction set, and the at least one instruction, at least one program, the code set or the instruction set is loaded and executed by a processor to implement a customer profile digital management method based on mining technology.

[0043] (III) Beneficial effects

[0044] Compared with the prior art, the technical solution provided by the present invention has the following beneficial effects:

[0045] 1. By using data mining technology to automatically process customer physical and virtual archive data, including feature extraction, classification and filtering, the time and cost of manual operations are greatly reduced, the speed and efficiency of data processing are improved, and the one-way fusion of physical features and virtual features is carried out. The virtual data is effectively integrated into the physical archive using data mining technology, and the correlation of archive data from different sources and forms of expression is constructed to provide users with more comprehensive and in-depth customer archive information, provide high-quality archive data, ensure the reliability of information, and reduce the risks caused by data errors or omissions.

[0046] 2. By defining index items, relevance scoring, data quality assessment and time range filtering methods, and combining data mining technology to finely filter the retrieved data, users can quickly obtain archival data that is highly matched to their needs, reducing the interference of irrelevant information and improving data retrieval accuracy.

[0047] 3. By customizing the identification framework, index items and filtering criteria according to the different business needs of users, it supports the processing of various types of archival data. It has high adaptability and can flexibly respond to the archival management needs of different fields and scales, and can still maintain efficient operation as the amount of data increases. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0049] Figure 1 It is a flow chart of the method for digital management of customer files based on mining technology in the present invention;

[0050] Figure 2 It is a schematic diagram of the framework of the customer archive digital management system based on mining technology in the present invention;

[0051] Figure 3 It is a state diagram of the optical character recognition waveform in the present invention.

[0052] The numbers in the figure represent: 1. Management module; 2. Framework construction module; 3. Scan conversion module; 4. Entity feature extraction module; 5. Criticality sorting module; 6. Virtual retrieval module; 7. Demand filtering module; 8. One-way fusion module; 9. Storage module. DETAILED DESCRIPTION

[0053] 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 ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0054] The present invention will be further described below in conjunction with the embodiments.

[0055] Example

[0056] The customer file digital management method based on mining technology in this embodiment is as follows: Figure 1 As shown, the following steps are included:

[0057] Step 1: Based on the goal and scope of archive digitization, define the archive types, fields and identification features that need to be identified, and complete the construction of the identification framework; based on the standards and regulatory requirements of specific industries, as well as the types, formats and importance of archives as references;

[0058] Step 2: Scan the physical archive according to the preset recognition framework, obtain the scanned image, pre-process it, and perform OCR character conversion to obtain the physical data; the scanning resolution, color mode and other parameters should be set according to the recognition framework, which will help improve the accuracy of subsequent processing steps and reduce errors to ensure the quality and integrity of the digital archive;

[0059] Step 3: Extract a number of entity features from entity data through natural language processing technology and group them according to established classification standards; identify relevant information from images, texts, and even structured data, and group them according to established classification standards, which will facilitate subsequent retrieval and indexing;

[0060] Step 4: Sort the grouped features by their criticality, and prioritize the highly critical features for indexable item definition;

[0061] Step 5: Obtain several indexable items and search them on the designated searchable platform, match them in the defined index items according to the user's query conditions, and extract relevant virtual archive data;

[0062] The process of extracting relevant virtual archive data is:

[0063] Select relevant items from a set of predefined index items based on the user's query conditions;

[0064] Use defined query conditions and index items to search for data on the selected searchable platform;

[0065] From the search results returned by the platform, filter by analyzing the matching degree with the user's query conditions, the release time of the data and the relevance of the data;

[0066] Organize and provide feedback on the screened virtual archive data;

[0067] Users can quickly obtain archival data that is highly matched to their needs, which reduces the interference of irrelevant information, improves the accuracy of data retrieval, provides high-quality archival data, ensures the reliability of information, and reduces the risks caused by data errors or omissions;

[0068] Step 6: Preset user demand indicators of time range, data quality and relevance score, and filter virtual archive data based on demand indicators combined with data mining technology;

[0069] The data mining process is:

[0070] Acquire virtual archive data within a specific time window and verify the quality of virtual archive data through data cleaning techniques and data verification rules;

[0071] Build a machine learning model to analyze keyword matching, content relevance, and user behavior data factors, obtain the matching degree between the current virtual data and the query purpose, and give priority to highly matching virtual data;

[0072] Step 7: From the filtered virtual data, extract the virtual features related to the original entity profile and perform one-way fusion with the extracted entity features. The fused features are output as the overall features of the customer profile. The attributes of the virtual features include: structured data, semantic information, or digital content corresponding to the entity profile. The virtual features will enhance or supplement the entity features without affecting the integrity of the original entity profile, thereby enhancing the richness and relevance of the profile data and providing users with more comprehensive and in-depth customer profile information.

[0073] Compared with existing technologies, by combining multiple data mining technologies, efficient, accurate, flexible and secure digital management of customer files is achieved. These advantages not only improve the overall efficiency and quality of file management, but also provide users with an intelligent and scalable file management solution.

[0074] In other aspects, this embodiment also provides a customer file digital management device based on mining technology, such as Figure 2 As shown, including:

[0075] Management module 1, as the core control of the archive management terminal, edits and sends control instructions to each functional module;

[0076] Framework building module 2 is used to identify the types, fields and features of archives that need to be digitized and establish an identification framework based on the goals and scope of archive digitization;

[0077] The scanning conversion module 3 is used to scan the physical archive or material according to the recognition framework, generate a scanned image, and convert the image into textual physical data;

[0078] Entity feature extraction module 4, used to extract entity features from text entity data, and classify and group the entity features;

[0079] A criticality ranking module 5 is used to rank the extracted entity features by criticality, and to recommend key features first. The features recommended first are defined as indexable items.

[0080] The virtual search module 6 is used to select relevant items from predefined indexable items according to the user's query conditions, perform data retrieval, filter search results by analyzing the matching degree with the query conditions, the release time of the data and the relevance, and extract the related virtual archive data;

[0081] Demand filtering module 7, used to filter virtual archive data according to user demand indicators, build a machine learning model to analyze the matching degree between virtual data and query purpose, and give priority to highly matching virtual data;

[0082] A one-way fusion module 8 is used to one-way fuse the highly matched virtual data with the extracted entity features and output the fused total features;

[0083] The storage module 9 is used as a data storage terminal to receive, transmit and back up interactive data, collected data and analyzed data in the cloud.

[0084] As a preferred implementation in this embodiment, Figure 2 As shown, the management module 1 is interactively connected with the framework construction module 2, the scan conversion module 3 and the storage module 9 through a wireless network, the scan conversion module 3 is interactively connected with the entity feature extraction module 4 through a wireless network, the entity feature extraction module 4 is interactively connected with the criticality sorting module 5 through a wireless network, the criticality sorting module 5 is interactively connected with the virtual retrieval module 6 through a wireless network, the virtual retrieval module 6 is interactively connected with the demand filtering module 7 through a wireless network, the demand filtering module 7 is interactively connected with the one-way fusion module 8 through a wireless network, and the one-way fusion module 8 is interactively connected with the storage module 9 through a wireless network.

[0085] During the specific implementation, the control instruction is sent by the management module 1, the scanning standard and the recognition framework are predefined by the framework construction module 2, the data of the entity archive is acquired by the scanning conversion module 3, the entity feature extraction module 4 extracts the entity feature of the entity archive data, the criticality sorting module 5 sorts the entity features by criticality, the virtual retrieval module 6 extracts virtual features based on high-criticality entity features, the demand filtering module 7 filters the obtained feature data based on the acquisition requirements, the virtual data is fused and matched with the entity data by the one-way fusion module 8 as the total feature of the user, and the data is recorded by the storage module 9.

[0086] In this embodiment, an OCR character conversion process is provided, which divides the scanned image into recognizable independent units, recognizes the information of each independent unit through an image information recognition method, performs parameter compensation, and compares and analyzes the data in a preset standard archive information library, determines the encoding content of the information based on the analysis results, and stores it in the database in a specified format. OCR technology can greatly improve the speed of file digitization processing, and its recognition quality is much higher than traditional recognition methods and manual entry methods. In the optical recognition process, its measurement waveform is as follows: Figure 3 As shown in the figure, the effect of this recognition method is directly related to the grayscale of the original file. If there is interference information during the acquisition process, a large number of errors will appear in the acquisition results. Therefore, parameter compensation is performed;

[0087] During the parameter compensation process, based on the recognition results of the independent unit information, the acquisition accuracy of the original file is controlled. The specific calculation process is as follows:

[0088]

[0089] In the formula, V1 and V2 represent the two boundary values ​​of the recognition cursor, V o represents the median value of the identification cursor, k represents the calculation constant, a represents the compensation parameter, S i represents the relative measurement result of archive image recognition, g represents the standard grayscale of paper archives, and by optimizing the optical character recognition process, the authenticity and accuracy of the archive collection results are guaranteed, and the collection results are used as the data basis in the archive management process.

[0090] As a preferred implementation manner of this embodiment, an electronic device is provided, which includes a processor and a memory, wherein the memory stores at least one instruction, at least one program, a code set or an instruction set, and the at least one instruction, at least one program, a code set or an instruction set is loaded and executed by the processor to implement a customer profile digital management method based on mining technology.

[0091] As another preferred implementation manner of this embodiment, a computer-readable storage medium stores at least one instruction, at least one program, a code set or an instruction set, and the at least one instruction, at least one program, a code set or an instruction set is loaded and executed by a processor to implement a customer profile digital management method based on mining technology.

[0092] In summary, the present invention integrates data from different sources, conducts comprehensive analysis, and provides a more comprehensive customer portrait. Through data mining, enterprises can better understand customer needs and preferences, thereby providing personalized services and product recommendations, improving customer satisfaction and loyalty, and can process new data in real time and quickly update customer profiles, so that enterprises can quickly respond to market changes and customer needs;

[0093] Automated data mining and analysis can reduce the time and cost of manual management and improve the efficiency of customer profile management. By analyzing customer behavior and feedback, companies can continuously optimize user experience and improve customer interaction and engagement.

[0094] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A customer file digital management method based on mining technology, characterized in that: The following steps are involved: Step 1: Based on the goal and scope of archive digitization, define the archive types, fields and identification features that need to be identified, and complete the construction of the identification framework; Step 2: Scan the physical file according to the preset recognition framework, obtain the scanned image, perform OCR character conversion after pre-processing, and obtain the physical data; Step 3: Extract a number of entity features from the entity data through natural language processing technology and group them according to the established classification criteria; Step 4: Sort the grouped features by their criticality, and prioritize the highly critical features for indexable item definition; Step 5: Obtain several indexable items and search them on the designated searchable platform, match them in the defined index items according to the user's query conditions, and extract relevant virtual archive data; Step 6: Preset user demand indicators of time range, data quality and relevance score, and filter virtual archive data based on demand indicators combined with data mining technology; Step 7: From the filtered virtual data, extract the virtual features related to the original entity profile and perform one-way fusion with the extracted entity features. The fused features are output as the total features of the customer profile.

2. The method for digital management of customer files based on mining technology according to claim 1 is characterized in that: The process of extracting relevant virtual archive data in step 5 is as follows: Select relevant items from a set of predefined index items based on the user's query conditions; Use defined query conditions and index items to search for data on the selected searchable platform; From the search results returned by the platform, filter by analyzing the matching degree with the user's query conditions, the release time of the data and the relevance of the data; The screened virtual archive data will be sorted and fed back.

3. The method for digital management of customer files based on mining technology according to claim 1 is characterized in that: During the OCR character conversion process in step 2, the scanned image is divided into recognizable independent units, and the information of each independent unit is identified by an image information recognition method, parameter compensation is performed, and the data is compared and analyzed with the data in a preset standard archive information library. The coded content of the information is determined based on the analysis results and stored in the database in a specified format.

4. The method for digital management of customer files based on mining technology according to claim 3 is characterized in that: In the parameter compensation process, based on the recognition result of the independent unit information, the acquisition accuracy of the original file is controlled. The specific calculation process is: In the formula, V1 and V2 represent the two boundary values ​​of the recognition cursor, V o represents the median value of the identification cursor, k represents the calculation constant, a represents the compensation parameter, S i represents the relative measurement result of archival image recognition, and g represents the standard grayscale of paper archives.

5. The method for digital management of customer files based on mining technology according to claim 1 is characterized in that: The data mining process in step 6 is as follows: Acquire virtual archive data within a specific time window and verify the quality of virtual archive data through data cleaning techniques and data verification rules; Build a machine learning model to analyze keyword matching, content relevance, and user behavior data factors, obtain the degree of matching between the current virtual data and the query purpose, and give priority to highly matching virtual data.

6. The method for digital management of customer files based on mining technology according to claim 1 is characterized in that: The attributes of the virtual features in step 7 include: structured data, semantic information or digital content corresponding to the physical archive.

7. A digital management device for customer files based on mining technology, the device is an implementation device of the digital management method for customer files based on mining technology according to any one of claims 1 to 6, characterized in that: include: Management module (1), as the core control of the archive management terminal, edits and sends control instructions to each functional module; The framework building module (2) is used to identify the types, fields and features of archives that need to be digitized according to the goals and scope of archive digitization and to establish an identification framework; A scanning conversion module (3) is used to scan the physical archive or material according to the recognition framework, generate a scanned image, and convert the image into textual physical data; An entity feature extraction module (4) is used to extract entity features from text entity data, and classify and group the entity features; A criticality ranking module (5) is used to rank the extracted entity features by criticality, and to recommend key features first. The features recommended first are defined as indexable items. A virtual search module (6) is used to select relevant items from predefined indexable items according to user query conditions, perform data search, filter search results by analyzing the matching degree with the query conditions, data release time and relevance, and extract related virtual archive data; A demand filtering module (7) is used to filter virtual archive data according to user demand indicators, build a machine learning model to analyze the matching degree between virtual data and query purpose, and give priority to highly matching virtual data; A one-way fusion module (8) is used to one-way fuse the highly matched virtual data with the extracted entity features and output the fused total features; The storage module (9) is used as a data storage terminal to receive, transmit and back up interactive data, collected data and analyzed data in the cloud.

8. The customer file digital management device based on mining technology according to claim 7 is characterized in that: The management module (1) is interactively connected with the framework construction module (2), the scan conversion module (3) and the storage module (9) via a wireless network; the scan conversion module (3) is interactively connected with the entity feature extraction module (4) via a wireless network; the entity feature extraction module (4) is interactively connected with the criticality sorting module (5) via a wireless network; the criticality sorting module (5) is interactively connected with the virtual retrieval module (6) via a wireless network; the virtual retrieval module (6) is interactively connected with the demand filtering module (7) via a wireless network; the demand filtering module (7) is interactively connected with the one-way fusion module (8) via a wireless network; and the one-way fusion module (8) is interactively connected with the storage module (9) via a wireless network.

9. An electronic device, characterized in that: The electronic device includes a processor and a memory, wherein the memory stores at least one instruction, at least one program, a code set or an instruction set, and the at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by the processor to implement the method as described in any one of claims 1-6.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores at least one instruction, at least one program, a code set or an instruction set, and the at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by a processor to implement the method described in any one of claims 1 to 6.

Images