Data mining method based on enterprise innovation method library

Abstract

The invention belongs to the technical field of enterprise management, and particularly relates to a data mining method based on an enterprise innovation method library, which comprises the following steps of: 1, establishing an enterprise data pool platform, and constructing a total innovation method library in the enterprise data pool platform; 2, building a data transmission network; 3, constructing a data mining module at the enterprise side, and completing scientific and technological innovation data mining; 4, constructing a data communication module at the enterprise side; 5, the enterprise terminal A sends a data retrieval demand to the total innovation method library; 6, after the general innovation method library is retrieved, the data owner enterprise terminal B is fed back; 7, the enterprise terminal A sends a data acquisition request to the enterprise terminal B; 8, after the enterprise terminal B completes identity verification, the data are encrypted and then sent to the enterprise terminal A; and step 9, all enterprise terminals synchronously record the data transmission process. The scientific and technological innovation data of the same industry can be queried and retrieved more efficiently and conveniently on the management enterprise.

Description

Technical Field

[0001] The present invention belongs to the technical field of enterprise management, and in particular relates to a data mining method based on an enterprise innovation method library. Background Art

[0002] In the process of enterprise management, it is often necessary to obtain innovation data in real time. In the process of data mining, it is necessary not only to mine the innovation data of the enterprise, but also to mine the innovation data of the same industry. When mining one's own innovation data, one can accurately know what technologies the enterprise has. When mining the innovation data of the same industry, one can achieve patent avoidance and patent navigation, and guide one's own scientific research direction at the same time, and finally promote the scientific and technological development of the entire industry. However, there are currently two ways to obtain scientific and technological data in the same industry. The first way is to search from data that has been made public, such as searching for the same type of data from a large number of patent websites. The second way is to share data and transmit data between enterprises, such as building a data pool and a sharing platform, and completing data transmission under the established sharing contract. This requires both enterprises to have information systems; and both parties need to develop interfaces, which is affected by the development cost cycle; or the most traditional way is to manually communicate through emails, telephones, instant tools, etc. to obtain corresponding data; Therefore, in the existing technology, there is a problem of more troublesome acquisition of innovation data in the process of enterprise management. For this reason, we propose a data mining method based on the enterprise innovation method library. Summary of the invention

[0003] The purpose of the present invention is to provide a data mining method based on an enterprise innovation method library. First, an enterprise data pool platform is built, and a total innovation method library is constructed in the enterprise data pool platform. Then, when one of the enterprise ends needs to obtain data, the enterprise end A as the data demander sends a data retrieval demand to the total innovation method library; the total innovation method library completes the retrieval according to the data retrieval demand, and feeds back the enterprise end B corresponding to the data owner of the required data; the enterprise end A sends a data acquisition request to the enterprise end B; after the enterprise end B completes the identity authentication, it retrieves the corresponding data from its own sub-innovation method library, and sends it to the enterprise end A after encryption; all enterprise ends synchronously record the data transmission process, so that in the process of enterprise management, the acquisition of innovation data in the same industry is faster, and the data transmission is decentralized, with higher security performance.

[0004] The technical solution adopted by the present invention is as follows: The data mining method based on the enterprise innovation method library includes the following steps: Step 1: Build an enterprise data pool platform, which is connected to several enterprise terminals and builds a total innovation method library in the enterprise data pool platform; Step 2: Build a data transmission network so that each enterprise end can communicate with each other; Step 3: Build a data mining module on the enterprise side. The data mining module mines the technological innovation data of the enterprise side, builds a sub-innovation method library, and uploads all the data in the sub-innovation method library to the overall innovation method library; Step 4: Build a data communication module on the enterprise side; Step 5: Enterprise A, as a data demander, sends a data retrieval request to the general innovation method library; Step 6: The total innovation method library completes the search according to the data search requirements and provides feedback to the enterprise end B corresponding to the data owner of the required data; Step 7: Enterprise A sends a data acquisition request to enterprise B; Step 8: After completing identity authentication, enterprise B retrieves the corresponding data from its own sub-innovation method library and sends it to enterprise A after encryption; Step 9: All enterprise terminals synchronously record the data transmission process.

[0005] Preferably, in step 1, the enterprise does not need to be verified when applying to connect to the data pool platform. It only needs to complete the construction of its own sub-innovation method library and upload all the data in its own sub-innovation method library to the overall innovation method library. Each enterprise has a private identity number.

[0006] Preferably, in step 2, each enterprise end is communicatively connected to the enterprise data pool platform, and each enterprise end is also communicatively connected to other enterprise ends that are commonly connected to the enterprise data pool platform.

[0007] Preferably, the data mining process in step 3 specifically includes the following steps: Step 301: Collect and clean the data of the enterprise end; Step 302: then intelligently classify the data; Step 303: Arrange and compile the data.

[0008] Preferably, step 301 specifically includes that the data collection link is used for the platform to obtain multi-source intelligence data, including customized collection of local open source data, and supports multimedia data formats, including text, pictures, web pages, audio, and video files; The data cleaning phase is used to deduplicate files, extract text, segment text, remove stop words, and perform structural preprocessing on local unstructured databases; file deduplication is achieved by comparing file URL links and HD5 value attribute values.

[0009] Preferably, in step 302, the intelligent classification link is used to make classification predictions for different topics on the structured vector data after automatic cleaning, and specifically uses expert rules and machine learning modes to jointly form intelligent classification.

[0010] Preferably, in step 303, the data is sorted and compiled according to the compilation report template, and the compilation report module is used to automatically generate a scientific and technological intelligence compilation report on a certain topic according to the topic and template set in the application layer foreground.

[0011] Preferably, in step 4, the enterprise side constructs a data communication module including an identity authentication unit and a data encryption transmission unit. The identity authentication module is mainly used to verify the identity, and the data encryption transmission unit is used to realize encrypted transmission of data.

[0012] Preferably, in step 5, the data in the total innovation method library cannot be viewed or downloaded, and the total innovation method library is only used to retrieve the owner of the data required by the enterprise end A as the data demander.

[0013] Preferably, in step 7, enterprise side A sends a data acquisition request to enterprise side B, including the required data summary, its own identity number and the encrypted private key generated by enterprise side A. In step 8, the identity authentication module of enterprise side B verifies whether it belongs to the enterprise data pool platform based on the identity number of enterprise side A, and determines whether to send data to enterprise side A. After the verification is completed and the data is agreed to be sent, the data encryption transmission unit uses the encryption key of enterprise side A to encrypt the required data, and sends the encrypted ciphertext directly to enterprise side A. Enterprise side A then completes data decryption based on the encrypted public key.

[0014] The technical effects achieved by the present invention are: In the present invention, an enterprise data pool platform is built, the enterprise data pool platform is connected to several enterprise terminals, and a general innovation method library is constructed in the enterprise data pool platform, and then a data transmission network is constructed so that each enterprise terminal is connected to each other in communication; then, a data mining module is constructed on the enterprise terminal, and the data mining module mines the scientific and technological innovation data of the enterprise terminal. In the data mining process, a sub-innovation method library is built, and the sub-innovation method library on the enterprise terminal includes images, operation videos, instructions for use and various research project data of the newly developed equipment, and finally realizes data including images, text, video, audio and various types of data, but there is a general name for the data, such as a method of using a certain machine, a three-dimensional model of a certain machine, and all the data in the sub-innovation method library are uploaded to the general innovation method library, and the construction of the enterprise data pool platform is completed. After the construction of the enterprise data pool platform is completed, it is more convenient to obtain innovation data in the process of enterprise management.

[0015] In the present invention, by setting up a total innovation method library, although the total innovation method library stores data related to scientific and technological innovation of all enterprise ends, any enterprise end cannot download data from the total innovation method library. The total innovation method library can be regarded as a retrieval terminal, which is only used to inform the data demander which enterprise end owns the data required by the data demander. The two enterprise ends need to connect with each other to complete point-to-point data encryption transmission, so as to avoid third parties from viewing the data, thereby greatly ensuring the security of data transmission. Since data transmission is point-to-point transmission through any two enterprise ends, decentralization is achieved, which ensures the efficiency of data transmission while ensuring the security of data transmission.

[0016] In the present invention, enterprise end A sends a data acquisition request to enterprise end B, wherein the data acquisition request includes a summary of the required data, the identity number of enterprise end A itself, and the encrypted private key generated by enterprise end A. At this time, enterprise end A also retains the generated encrypted public key. After successful verification and agreement to share data, the data encryption transmission unit of enterprise end B uses the encryption key of enterprise end A to encrypt the required data, and sends the encrypted ciphertext directly to enterprise end A. Enterprise end A then completes data decryption based on the encrypted public key. Through the asymmetric encryption algorithm, data transmission between any two enterprise ends is made more secure.

[0017] In summary, we build an enterprise data pool platform, complete the construction of the overall innovation method library, connect several enterprise terminals at the same time, each enterprise terminal completes point-to-point connection with other enterprise terminals, completes scientific and technological innovation data mining within the enterprise terminal, and builds a sub-innovation method library, so that in the management of enterprises, it is more efficient and convenient to query and retrieve scientific and technological innovation data in the same industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a flow chart of the data mining method based on the enterprise innovation method library of the present invention; Figure 2 It is a system block diagram of an enterprise data pool platform in a data mining method based on an enterprise innovation method library of the present invention; Figure 3 It is a system block diagram of the enterprise end in the data mining method based on the enterprise innovation method library of the present invention; Figure 4 It is a flowchart of mining scientific and technological innovation data within the enterprise in step 3 of the data mining method based on the enterprise innovation method library of the present invention; Figure 5 It is a flow chart of data sharing between enterprise side A and enterprise side B in the data mining method based on the enterprise innovation method library of the present invention. DETAILED DESCRIPTION

[0019] In order to make the purpose and advantages of the present invention more clearly understood, the present invention is specifically described below in conjunction with embodiments. It should be understood that the following text is only used to describe one or several specific embodiments of the present invention, and does not strictly limit the scope of protection of the specific claims of the present invention.

[0020] like Figure 1 As shown in FIG. 1 , the data mining method based on the enterprise innovation method library includes the following steps: Step 1: Build an enterprise data pool platform, which is connected to several enterprise terminals and builds a total innovation method library in the enterprise data pool platform; Step 2: Build a data transmission network so that each enterprise end can communicate with each other; Step 3: Build a data mining module on the enterprise side. The data mining module mines the technological innovation data of the enterprise side, builds a sub-innovation method library, and uploads all the data in the sub-innovation method library to the overall innovation method library; Step 4: Build a data communication module on the enterprise side; Step 5: Enterprise A, as a data demander, sends a data retrieval request to the general innovation method library; Step 6: The total innovation method library completes the search according to the data search requirements and provides feedback to the corresponding enterprise end B as the data owner of the required data; Step 7: Enterprise A sends a data acquisition request to enterprise B; Step 8: After completing identity authentication, enterprise B retrieves the corresponding data from its own sub-innovation method library and sends it to enterprise A after encryption; Step 9: All enterprise terminals synchronously record the data transmission process.

[0021] In the present invention, by constructing an enterprise data pool platform and building a general innovation method library in the enterprise data pool platform, it is possible to organize similar enterprises, and in realizing enterprise innovation management, the scientific research direction of the entire industry can be more intuitively seen, and there is more precise guidance in R&D scientific and technological innovation direction planning, patent technology avoidance and patent navigation.

[0022] like Figure 2 As shown, in step 1, the enterprise does not need to be verified when applying to connect to the data pool platform. It only needs to complete the construction of its own sub-innovation method library and upload all the data in its own sub-innovation method library to the overall innovation method library. Each enterprise has a private identity number.

[0023] In the present invention, the private identity number of the enterprise side is used as the unique identity information authenticated by the enterprise side on the data pool platform. The identity numbers of different enterprise sides are all different. The private identity number of the enterprise side makes the enterprise side management safer.

[0024] like Figure 3 As shown, in step 2, each enterprise end is connected to the enterprise data pool platform for communication, and each enterprise end is also connected to other enterprise ends that are connected to the enterprise data pool platform for communication.

[0025] In the present invention, each enterprise end is connected to the data pool platform to achieve communication with the data pool platform. At the same time, each enterprise end is also connected to all other ends, so that two-way communication and data transmission can be achieved between any two enterprise ends.

[0026] like Figure 4 As shown, the data mining process in step 3 specifically includes the following steps: Step 301: Collect and clean the data of the enterprise end; Step 302: then intelligently classify the data; Step 303: Arrange and compile the data.

[0027] Step 301 specifically includes that the data collection phase is used by the platform to obtain multi-source intelligence data, including customized collection of local open source data, and supports multimedia data formats, including text, pictures, web pages, audio, and video files; It should be noted that the optional parameters supported in the design include: target source, that is, the website URL or local path to be collected, and the Chinese and English names, language type, and website attributes can be entered at the same time; mining level, that is, the access depth under the URL or root path; external domain permissions, that is, whether to access data outside the domain name or root path through links or shortcuts within the site, and file types, which can filter pdf, doc, html, jpg, and mp4; number of threads, that is, the number of multi-threads used in the collection task; triggering methods, including manual triggering of collection and automatic collection at a specified time; the local collection function also supports one-click import and structured extraction of various files under a given local path and databases with specified names; the various types of raw data obtained above are stored in the local raw unstructured database, where files are stored on the local disk and file attributes are stored in the database table; for all raw data, the enterprise end can browse, view, manually modify, and batch download through conditional retrieval at the front end; The data cleaning stage is used for file deduplication, text extraction, text segmentation, stop word removal, and structural preprocessing functions of the local unstructured database; file deduplication is achieved by comparing the URL link and HD5 value attribute value of the file; text extraction is to convert multimedia files into text files, that is, optical characters rather than text characters in pdf, word, html, and jpg files are extracted into txt text through tag structure and optical character recognition methods, audio files are subjected to speech recognition to obtain txt text, and subtitles and audio are extracted and converted into txt text after frame extraction of video files; text segmentation is to perform n-gram segmentation on the text, and remove stop words according to the commonly used stop word library or custom stop word library set in the system to obtain a vector file; after data collection and data cleaning, the system realizes the acquisition of structured vectors of the original data.

[0028] In step 302, the intelligent classification link is used to make classification predictions for different topics for the structured vector data after automatic cleaning. Specifically, the two modes of expert rules and machine learning are used to form intelligent classification, which makes the disordered and non-topic database orderly according to the user's attention, and is the core part of the platform. In order to achieve better classification prediction performance, the present invention proposes to use the two modes of expert rules and machine learning to form an intelligent classification engine; the essence of expert rule classification is regular operation, that is, experts set explicit rules based on industry experience or statistical results. For each intelligence theme classification and its subclasses, the basic keyword library under the theme is defined through front-end operations, and the classification rules are clearly configured using AND or non-Boolean logic and word set / bag of words model. The background will automatically generate a regular classifier according to this rule, that is, an expert rule classifier; machine learning classification is based on the machine learning samples provided by the user, based on one or several machine learning algorithms including naive Bayes, support vector machine, random forest, Bayesian network, K nearest neighbor, and automatically trained by the background to generate a classifier code, that is, a machine learning classifier.

[0029] In step 303, the data is sorted and compiled according to the compilation report template. The compilation report module is used to automatically generate a scientific and technological intelligence compilation report on a certain topic according to the topic and template set by the application layer foreground. The topic, structure, format, and content of the report can be customized by the user at the foreground, and the granularity can be refined to each paragraph. The definition of paragraph content can adopt clear expert rules. When there is intelligence text in the classified intelligence database that matches the rule logic, the intelligence text will be compiled; or machine learning can be used, and the user provides example text for each report.

[0030] It should be added that for each enterprise side, after data mining, the sub-innovation method library of the enterprise side includes images, operation videos, instructions for use and various research project data of the newly developed equipment, and ultimately includes images, text, video, audio and various types of data, but the data has a general name, such as the method of using a certain machine or the three-dimensional model of a certain machine.

[0031] like Figure 3 As shown, in step 4, the enterprise side constructs a data communication module including an identity authentication unit and a data encryption transmission unit. The identity authentication module is mainly used to verify the identity, and the data encryption transmission unit is used to realize encrypted transmission of data.

[0032] In the present invention, by setting an identity authentication unit and a data encryption transmission unit, identity authentication and data encryption transmission are completed during data transmission between two enterprise ends, thereby preventing data from being maliciously viewed.

[0033] In step 5, the data in the total innovation method library cannot be viewed or downloaded. The total innovation method library is only used to retrieve the owner of the data required by enterprise A as a data demander.

[0034] By setting up a general innovation method library, although the general innovation method library stores data related to scientific and technological innovation of all enterprise ends, no enterprise end can download data from the general innovation method library. The general innovation method library can be regarded as a retrieval terminal, which is only used to inform the data demander which enterprise end owns the required data. The two enterprise ends need to connect with each other to complete point-to-point data encryption transmission to prevent third parties from viewing the data, which greatly ensures the security of data transmission. Since data transmission is point-to-point transmission through any two enterprise ends, decentralization is achieved, which ensures data transmission efficiency while ensuring the security of data transmission.

[0035] like Figure 5 As shown, in step 7, enterprise side A sends a data acquisition request to enterprise side B, including the required data summary, its own identity number and the encryption private key generated by enterprise side A. In step 8, the identity authentication module of enterprise side B verifies whether it belongs to the enterprise data pool platform based on the identity number of enterprise side A. After the verification is completed, the data encryption transmission unit uses the encryption key of enterprise side A to encrypt the required data, and sends the encrypted ciphertext directly to enterprise side A. Enterprise side A completes data decryption based on the encryption public key.

[0036] It should be added that the data encryption transmission method adopted by the data encryption transmission unit is an asymmetric encryption algorithm. Asymmetric encryption requires two keys: a public key and a private key. The public key and the private key are a pair. If the data is encrypted with the public key, it can only be decrypted with the corresponding private key. If the data is encrypted with the private key, it can only be decrypted with the corresponding public key. Because different keys are used for encryption and decryption, it is called asymmetric encryption; the confidentiality of the asymmetric encryption algorithm is good, it eliminates the need for end users to exchange keys, the algorithm strength is complex, and the security depends on the algorithm and the key. However, due to its complex algorithm, the encryption and decryption speed is not as fast as the symmetric encryption and decryption speed. There is only one key in the symmetric encryption system, and it is non-public. If you want to decrypt, you have to let the other party know the key; so to ensure its security is to ensure the security of the key, and the asymmetric key system has two keys, one of which is public, so there is no need to transmit the other party's key like symmetric encryption. In this way, security is greatly improved.

[0037] It should be noted that during the identity authentication process, the first step is to determine whether the enterprise side is connected to the enterprise data pool platform, and secondly, whether the data that the enterprise as the data demander needs to view can be shared. For example, enterprise side B owns the data, but as confidential data, enterprise side B does not want to be viewed by enterprise side A. At this time, enterprise side B can provide feedback to enterprise side A, indicating that the data the other party needs to obtain is corporate confidential and cannot be viewed, or the two parties communicate on what conditions can be used to obtain confidential data, such as paying the corresponding fees.

[0038] It should be added that, for example, if the data that enterprise A wants to view is the data in enterprise A's own sub-innovation method library, similarly, enterprise A sends a data retrieval request to the overall innovation method library; the overall innovation method library completes the retrieval based on the data retrieval request, and feeds back that the data owner of the required data is enterprise A itself. At this time, enterprise A directly retrieves the corresponding data from its own sub-innovation method library.

[0039] like Figure 5 As shown, it is worth mentioning that during the data flow process, all enterprise terminals synchronously record the data transmission process to ensure the traceability of data transmission and further ensure the security of data transmission. Moreover, when an enterprise terminal needs to obtain certain data, it can not only send data retrieval requirements to the total innovation method library, but also directly retrieve its own data transmission records, and retrieve whether the data it needs has been shared through its own data transmission records. For shared data, it can directly obtain the enterprise terminal that is the data owner. For data that has not been shared, it will send data retrieval requirements to the total innovation method library to further improve the effect of sharing data.

[0040] It should be added that when one of the enterprise ends sends a data retrieval request to the overall innovation method library, the overall innovation method library completes the retrieval based on the data retrieval request and provides feedback to multiple enterprise ends corresponding to the data owners of the required data, indicating that multiple enterprise ends have relevant data. The enterprise end as the data demander can simultaneously send data acquisition requests to multiple data demanders who own the data.

[0041] The working principle of the present invention is: As shown in Figure 5, first: build an enterprise data pool platform, the enterprise data pool platform communicates with several enterprise terminals, and builds a general innovation method library in the enterprise data pool platform, and then builds a data transmission network so that each enterprise terminal can communicate with each other; then, build a data mining module on the enterprise side, the data mining module completes the mining of the scientific and technological innovation data of the enterprise side, and in the data mining process, build a sub-innovation method library. The sub-innovation method library on the enterprise side includes images, operation videos, instructions for use and various research project data of the newly developed equipment, and finally realizes data including images, text, video, audio and various types of data, but there is a general name for the data, such as the use method of a certain machine, the three-dimensional model of a certain machine, and all the data in the sub-innovation method library are uploaded to the general innovation method library, thus completing the construction of the enterprise data pool platform.

[0042] It has to be said that after completing the construction of the enterprise data pool platform, it is more convenient to obtain innovative data during the enterprise management process.

[0043] After building the enterprise data pool platform, for example, enterprise A among many enterprise ends wants to obtain a certain type of data, such as sweeping robot data. Enterprise A sends "sweeping robot" as a data summary to the general innovation method library to send a data retrieval requirement; the general innovation method library completes the retrieval according to the data retrieval requirement, and feedbacks enterprise B corresponding to the data owner of the required data, indicating that enterprise B has relevant data on "sweeping robot", but the general innovation method library will not display the specific data of enterprise B about "sweeping robot", and then enterprise A sends a data acquisition request to enterprise B, where the data acquisition request includes the required data summary "sweeping robot", enterprise A's own identity number and the encrypted private key generated by enterprise A. At this time, enterprise A also retains the generated encrypted public key, and the identity authentication module of enterprise B verifies whether it belongs to the enterprise A according to its identity number. Enterprise data pool platform, secondly, based on whether the data that the enterprise that needs to view can be shared, for example, enterprise B has the data, but as confidential data, enterprise B does not want it to be viewed by enterprise A. At this time, enterprise B can give feedback to enterprise A, expressing that the data the other party needs to obtain is corporate confidential and cannot be viewed, or the two parties communicate on what conditions can be used to obtain confidential data, such as paying corresponding fees. Finally, when enterprise B agrees to data sharing, the data encryption transmission unit of enterprise B uses the encryption key of enterprise A to encrypt the required data, and sends the encrypted ciphertext directly to enterprise A. Enterprise A completes data decryption based on the encrypted public key. Finally, all enterprise terminals synchronously record the data transmission process, recording, for example: enterprise A obtains "sweeping robot" related data from enterprise B, but also does not record the specific data content.

[0044] The above is only a preferred embodiment of the present invention. It should be noted that, for those skilled in the art, several improvements and modifications can be made without departing from the principles of the present invention, and these improvements and modifications should also be considered as the protection scope of the present invention. The structures, devices and operating methods not specifically described and explained in the present invention shall be implemented according to the conventional means in the art unless otherwise specified and limited.

Claims

1. A data mining method based on an enterprise innovation method library, characterized by: The following steps are involved: Step 1: Build an enterprise data pool platform, which is connected to several enterprise terminals and builds a total innovation method library in the enterprise data pool platform; Step 2: Build a data transmission network so that each enterprise end can communicate with each other; Step 3: Build a data mining module on the enterprise side. The data mining module mines the technological innovation data of the enterprise side, builds a sub-innovation method library, and uploads all the data in the sub-innovation method library to the overall innovation method library; Step 4: Build a data communication module on the enterprise side; Step 5: Enterprise A, as a data demander, sends a data retrieval request to the general innovation method library; Step 6: The total innovation method library completes the search according to the data search requirements and provides feedback to the enterprise end B corresponding to the data owner of the required data; Step 7: Enterprise A sends a data acquisition request to enterprise B; Step 8: After completing identity authentication, enterprise B retrieves the corresponding data from its own sub-innovation method library and sends it to enterprise A after encryption; Step 9: All enterprise terminals synchronously record the data transmission process.

2. The data mining method based on the enterprise innovation method library according to claim 1 is characterized by: In step 1, the enterprise does not need to be verified when applying to connect to the data pool platform. It only needs to complete the construction of its own sub-innovation method library and upload all the data in its own sub-innovation method library to the overall innovation method library. Each enterprise has a private identity number.

3. The data mining method based on the enterprise innovation method library according to claim 1 is characterized by: In step 2, each enterprise end is connected to the enterprise data pool platform for communication, and each enterprise end is also connected to other enterprise ends that are connected to the enterprise data pool platform for communication.

4. The data mining method based on the enterprise innovation method library according to claim 1 is characterized by: The data mining process in step 3 specifically includes the following steps: Step 301: Collect and clean the data of the enterprise end; Step 302: then intelligently classify the data; Step 303: Arrange and compile the data.

5. The data mining method based on the enterprise innovation method library according to claim 4 is characterized by: Step 301 specifically, the data collection phase is used for the platform to obtain multi-source intelligence data, including customized collection of local open source data, and supports multimedia data formats, including text, pictures, web pages, audio, and video files; The data cleaning phase is used to deduplicate files, extract text, segment text, remove stop words, and perform structural preprocessing on local unstructured databases; file deduplication is achieved by comparing file URL links and HD5 value attribute values.

6. The data mining method based on the enterprise innovation method library according to claim 5 is characterized by: In step 302, the intelligent classification link is used to make classification predictions for different topics on the structured vector data after automatic cleaning, and specifically uses expert rules and machine learning models to form intelligent classification.

7. The data mining method based on the enterprise innovation method library according to claim 6 is characterized by: In step 303, the data is sorted and compiled according to the compilation report template. The compilation report module is used to automatically generate a scientific and technological intelligence compilation report on a certain topic according to the topic and template set by the application layer foreground.

8. The data mining method based on the enterprise innovation method library according to claim 7 is characterized by: In step 4, the enterprise side constructs a data communication module including an identity authentication unit and a data encryption transmission unit. The identity authentication module is mainly used to verify the identity, and the data encryption transmission unit is used to realize encrypted transmission of data.

9. The data mining method based on the enterprise innovation method library according to claim 8 is characterized by: In step 5, the data in the total innovation method library cannot be viewed or downloaded. The total innovation method library is only used to retrieve the owner of the data required by enterprise A as a data demander.

10. The data mining method based on the enterprise innovation method library according to claim 8 is characterized by: In step 7, enterprise A sends a data acquisition request to enterprise B, including the required data summary, its own identity number and the encrypted private key generated by enterprise A. In step 8, the identity authentication module of enterprise B verifies whether it belongs to the enterprise data pool platform based on the identity number of enterprise A, and determines whether to send data to enterprise A. After the verification is completed and the data is agreed to be sent, the data encryption transmission unit uses the encryption key of enterprise A to encrypt the required data, and sends the encrypted ciphertext directly to enterprise A. Enterprise A then completes data decryption based on the encrypted public key.

Images