Information classification management system and method based on data analysis

Abstract

The invention discloses an information classification management system and method based on data analysis, and relates to the technical field of information data analysis. Historical classification records of non-structural data are collected, feature data sequences of all non-structural data files in a classification category are obtained, and according to numerical features of the feature data sequences, the non-structural data files in the classification category are classified; establishing a reference data sequence of a classification category, obtaining a target file, taking a feature data sequence of the target file as a target data sequence, recording the difference between the target data sequence and the reference data sequence as a reference difference, and calculating the difference of the same information attribute subsets in the target data sequence and the reference data sequence; and collecting information attribute subsets smaller than the reference difference, obtaining classification records including the target file from the historical classification records, calculating the classification weight of each information attribute in the target file according to the frequency of recording each information attribute as a feature information attribute, and outputting a key data list of the target file.

Description

Technical Field

[0001] The present invention relates to the technical field of information data analysis, and in particular to an information classification management system and method based on data analysis. Background Art

[0002] Unstructured information refers to data that does not conform to the traditional structured data model. It usually exists in the form of free text, images, audio, video, etc., without a fixed format and organizational structure. Unstructured data is common information in people's daily lives. Therefore, classifying such information that lacks a clear structure and organization is a common problem in the field.

[0003] In the prior art, information clustering or principal component analysis is often used to assist relevant personnel in classifying unstructured data. However, the accuracy of classification in the prior art depends on the quality of the data. If there are outliers or noise in the data, it may affect the analysis results. Therefore, it is necessary to further reduce the impact of data noise on the classification of unstructured information. Summary of the invention

[0004] The purpose of the present invention is to provide an information classification management system and method based on data analysis to solve the problems raised in the prior art.

[0005] To achieve the above object, the present invention provides the following technical solution: an information classification management method based on data analysis, the method comprising: Step S100: classifying the information of the non-structured data file, classifying and extracting the features of each type of information in the non-structured data file, and converting a non-structured data file into a corresponding feature data sequence; Step S200: Collect historical classification records of unstructured data, obtain characteristic data sequences of all unstructured data files in a classification category, and establish a reference data sequence of the classification category based on the numerical features of the characteristic data sequence; Step S300: obtaining a non-structured data file from the classification category as a target file, taking a characteristic data sequence of the target file as a target data sequence, and recording a difference between the target data sequence and a reference data sequence as a reference difference; Step S400: Calculate the difference between the same information attribute subsets in the target data sequence and the reference data sequence, collect the information attribute subsets that are smaller than the reference difference, and record them as characteristic information attributes; Step S500: Obtain classification records including the target file from historical classification records, calculate the classification weight of each information attribute in the target file according to the frequency of each information attribute being recorded as a characteristic information attribute, and output a key data list of the target file.

[0006] Furthermore, step S100 includes: Step S101: obtaining i types of information attributes in a target file, and forming the information attributes of the target file into an information attribute set, wherein DA: a1, a2, a3, ..., ai, respectively represent the first, second, third, ..., and i-th information attributes of the target file; Step S102: Obtain the file content corresponding to the information attribute set DA, extract the feature vector of the file content to obtain the content feature vector, collect all the content feature vectors of the target file, and obtain the content vector set DB, DB: b1, b2, b3, ... bi, representing the content feature vectors corresponding to the first, second, third, ... and i-th information attributes of the target file respectively; Step S103: The corresponding information attributes and file contents are grouped into information groups, and all information groups of the target file are collected to obtain a characteristic data sequence of the target file, wherein the i-th information group in the characteristic data sequence is (ai, bi).

[0007] Further, step S200 includes: Step S201: record a classification category including a target file in a certain historical classification record as a target category, obtain n non-structured data files in the target category, and obtain a feature data sequence corresponding to each non-structured data file; Step S202: extracting information attribute sets of each unstructured data file respectively, calculating the intersection of the information attribute sets to obtain a classification attribute set, and obtaining the total number of information attributes in the classification attribute set, which is recorded as q; Step S203: Obtain the number r of all information attributes included in the e-th unstructured data in the target category e , calculate the total number of attributes Q in the target category, ; Step 204: In the target category, the number of occurrences of the same information attribute in all feature data sequences is traversed, where the number of occurrences of the jth information attribute in the classification attribute set is recorded as att j , calculate the attribute weight α of the j-th information attribute j , α j =att j / Q; Step S205: Obtain m content feature vectors of the j-th information attribute in the classification category, and calculate the reference vector ref of the j-th information attribute j , , where k f represents the fth content feature vector among the m content feature vectors; This step reconstructs the vectors in the classification category, collects the attribute characteristics of the vectors in the category, and reduces the weight value of the vector corresponding to the infrequent attribute in the category; Step S206: Collect all information attributes in the target category and the reference vectors corresponding to each information attribute, and group the information attributes and the corresponding reference vectors into data groups to form a reference data sequence, wherein the g-th data group in the reference data sequence is (tri g , equal g ), tri g represents the information attribute in the g-th data group, g Indicates tri g The corresponding reference vector.

[0008] Furthermore, step S300 includes: Step S301: Obtain the classification attribute set P0, calculate the reference set P1, P1=P0∩DA, and obtain the reference vector corresponding to each information attribute in P1 from the reference data sequence; Step S302: Grouping the content feature vectors in the content vector set DB into a first vector group X of i dimensions i 1. The reference vectors corresponding to each information attribute in P1 are combined into a second vector group X of i dimensions i 2. Calculate the distance d0 between the first vector group and the second vector group, and record the reference distance as the difference between the target data sequence and the reference data sequence as the reference difference; The file features of the target file that are helpful for the classification operation are extracted during the classification process. In numerical terms, the difference between the sequence composed of several sub-features of the target file and the reference data sequence is smaller.

[0009] Furthermore, step S400 includes: Step S301: Obtain the classification attribute set P0, calculate the reference set P1, P1=P0∩DA, and obtain the reference vector corresponding to each information attribute in P1 from the reference data sequence; Step S302: Grouping the content feature vectors in the content vector set DB into a first vector group X of i dimensions i 1. The reference vectors corresponding to each information attribute in P1 are combined into a second vector group X of i dimensions i 2. Calculate the distance d0 between the first vector group and the second vector group, and record the reference distance as the difference between the target data sequence and the reference data sequence as the reference difference.

[0010] Furthermore, step S500 includes: Step S501: collect t times of historical classification records of the target file, collect the number of times each information attribute in the target file is recorded as a characteristic information attribute, where the number of times the i-th information attribute of the target file is recorded as a characteristic information attribute is mes i ; Step S502: Calculate the classification weight corresponding to each information attribute of the target file, where the classification weight of the i-th information attribute γi, γi=mes i / t; Step S503: setting a weight threshold w. When the classification weight of the information attribute is greater than the weight threshold w, the information attribute and content feature vector corresponding to the classification weight are collected and recorded in the key data list of the target file.

[0011] In order to better implement the above method, an information classification management system based on data analysis is also proposed, the system includes: A feature data sequence management module, a reference data sequence management module, a reference difference calculation module, a feature information attribute management module and a key data list management module, wherein the feature data sequence management module is used to extract information attributes and content vectors of non-structured data respectively and collect them to obtain feature data sequences, the reference data sequence management module is used to manage reference data sequences of classification categories, the reference difference calculation module is used to calculate reference differences, the feature information attribute management module is used to collect information attribute subsets that are smaller than reference differences, and the key data list management module is used to manage key data lists of non-structured data; Further, the feature data sequence management module includes: an information attribute management unit, a content feature management unit and a feature data sequence management unit, wherein the information attribute management unit is used to manage the information attribute set of the non-structured data, the content feature management unit is used to manage the content vector set of the non-structured data, and the feature data sequence management unit is used to manage the feature data sequence of the non-structured data; Further, the reference data sequence management module includes: a total attribute management unit, an attribute weight calculation unit, a reference vector calculation unit and a data collection unit, wherein the total attribute management unit is used to obtain the total number of attributes in the target category, the attribute weight calculation unit is used to calculate the attribute weight of the information attribute, the reference vector calculation unit is used to calculate the reference vector of the information attribute, and the data collection unit is used to collect the information attribute and the reference vector to obtain the reference data sequence; Further, the reference difference calculation module includes: a vector group collection unit and a difference calculation unit, wherein the vector group collection unit is used to collect the first vector group and the second vector group, and the difference calculation unit is used to calculate the difference between the vector groups; Furthermore, the characteristic information attribute management module includes: a subset management unit and an attribute collection unit, wherein the subset management unit is used to obtain a subset in the vector group and construct a comparison attribute set, and the attribute collection unit is used to collect the characteristic attribute set to obtain information attributes of the existing characteristic information attributes; Further, the reference difference calculation module includes: a vector group collection unit and a difference calculation unit, wherein the vector group collection unit is used to collect the first vector group and the second vector group, and the difference calculation unit is used to calculate the difference between the vector groups; The characteristic information attribute management module includes: a subset management unit and an attribute collection unit, wherein the subset management unit is used to obtain a subset in the vector group and construct a comparison attribute set, and the attribute collection unit is used to collect the characteristic attribute set to obtain information attributes of the existing characteristic information attributes.

[0012] Compared with the prior art, the beneficial effects of the present invention are: by collecting historical records of unstructured data, data features in the classification category are extracted, and the part of the data in the target file that plays a role in the classification is further analyzed. The parts of the target file that play a more important role in the classification process are extracted from several historical classification records to generate a key data list of the target file. In the subsequent comparison and classification of the target file, the interference of data noise in the target file is eliminated, thereby improving the efficiency of file feature comparison and file classification. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 This is a structural schematic diagram of an information classification management system based on data analysis according to the present invention; Figure 2 The present invention is a flowchart of an information classification management method based on data analysis. DETAILED DESCRIPTION

[0014] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only 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 creative work are within the scope of protection of the present invention.

[0015] Example: Figure 1 and Figure 2 As shown, the present invention provides a technical solution, an information classification management method based on data analysis: Step S100: classifying the information of the non-structured data file, classifying and extracting the features of each type of information in the non-structured data file, and converting a non-structured data file into a corresponding feature data sequence; Wherein, step S100 includes: Step S101: obtaining i types of information attributes in a target file, and forming the information attributes of the target file into an information attribute set, wherein DA: a1, a2, a3, ..., ai, respectively represent the first, second, third, ..., and i-th information attributes of the target file; Step S102: Obtain the file content corresponding to the information attribute set DA, extract the feature vector of the file content to obtain the content feature vector, collect all the content feature vectors of the target file, and obtain the content vector set DB, DB: b1, b2, b3, ... bi, representing the content feature vectors corresponding to the first, second, third, ... and i-th information attributes of the target file respectively; Step S103: The corresponding information attributes and file contents are grouped into information groups, and all information groups of the target file are collected to obtain a characteristic data sequence of the target file, wherein the i-th information group in the characteristic data sequence is (ai, bi).

[0016] Step S200: Collect historical classification records of unstructured data, obtain characteristic data sequences of all unstructured data files in a classification category, and establish a reference data sequence of the classification category based on the numerical features of the characteristic data sequence; Wherein, step S200 includes: Step S201: record a classification category including a target file in a certain historical classification record as a target category, obtain n non-structured data files in the target category, and obtain a feature data sequence corresponding to each non-structured data file; Step S202: extracting information attribute sets of each unstructured data file respectively, calculating the intersection of the information attribute sets to obtain a classification attribute set, and obtaining the total number of information attributes in the classification attribute set, which is recorded as q; Step S203: Obtain the number r of all information attributes included in the e-th unstructured data in the target category e , calculate the total number of attributes Q in the target category, ; Step 204: In the target category, the number of occurrences of the same information attribute in all feature data sequences is traversed, where the number of occurrences of the jth information attribute in the classification attribute set is recorded as att j , calculate the attribute weight α of the jth information attribute j , α j =att j / Q; Step S205: Obtain m content feature vectors of the j-th information attribute in the classification category, and calculate the reference vector ref of the j-th information attribute j , , where k frepresents the fth content feature vector among m content feature vectors; In the embodiment, there are 6 types of information attributes, 6 file features of unstructured data are extracted, and the file features are converted into numerical values, and the file features are represented by feature values, and the file features are arranged in order. When a data feature is not extracted for a certain information attribute, 0 is used as a placeholder; R1: (0, 3, 2, 3, 0, 1), R2: (2, 1, 1, 0, 4, 3), R3: (3, 0, 2, 1, 1, 0), R4: (2, 3, 0, 1, 3, 2), R5: (4, 2, 4, 1, 3, 1), R6 (3, 2, 2, 0, 2, 4); The attribute weights of information attributes are: 0.172, 0.172, 0.172, 0.138, 0.172, 0.172; The reference vector for this classification is: (2.408, 1.892, 1.892, 0.828, 2.236, 1.892); Step S206: Collect all information attributes in the target category and the reference vectors corresponding to each information attribute, and group the information attributes and the corresponding reference vectors into data groups to form a reference data sequence, wherein the g-th data group in the reference data sequence is (tri g , equal g ), tri g represents the information attribute in the g-th data group, g Indicates tri g The corresponding reference vector.

[0017] Step S300: obtaining a non-structured data file from the classification category as a target file, taking a characteristic data sequence of the target file as a target data sequence, and recording a difference between the target data sequence and a reference data sequence as a reference difference; Wherein, step S300 includes: Step S301: Obtain the classification attribute set P0, calculate the reference set P1, P1=P0∩DA, and obtain the reference vector corresponding to each information attribute in P1 from the reference data sequence; Step S302: Grouping the content feature vectors in the content vector set DB into a first vector group X of i dimensions i 1. The reference vectors corresponding to each information attribute in P1 are combined into a second vector group X of i dimensions i 2. Calculate the distance d0 between the first vector group and the second vector group, and record the reference distance as the difference between the target data sequence and the reference data sequence as the reference difference.

[0018] Step S400: Calculate the difference between the same information attribute subsets in the target data sequence and the reference data sequence, collect the information attribute subsets that are smaller than the reference difference, and record them as characteristic information attributes; Wherein, step S400 includes: Step S401: Obtain h content feature vectors from the first vector group to form an h-dimensional vector group X h 1. Gather the information attributes corresponding to the h content feature vectors to obtain the comparison attribute set U; Step S402: Obtain reference vectors corresponding to each information attribute in the comparison attribute set from the second vector group, and collect the reference vectors to form an h-dimensional vector group X h 2, Calculate the vector group X h 1 and X h 2 The distance d h ; Step S403: When d h When <d0, the comparison attribute set U is recorded as a characteristic attribute set, the content characteristic vectors in the first vector group are traversed to obtain several characteristic attribute sets, the intersection of the several characteristic attribute sets is calculated, and all information attributes included in the intersection are recorded as characteristic information attributes; Step S500: Obtain classification records including the target file from historical classification records, calculate the classification weight of each information attribute in the target file according to the frequency of each information attribute being recorded as a characteristic information attribute, and output a key data list of the target file; Wherein, step S500 includes: Step S501: collect t times of historical classification records of the target file, collect the number of times each information attribute in the target file is recorded as a characteristic information attribute, where the number of times the i-th information attribute of the target file is recorded as a characteristic information attribute is mes i ; Step S502: Calculate the classification weight corresponding to each information attribute of the target file, where the classification weight of the i-th information attribute γi, γi=mes i / t; Step S503: setting a weight threshold w. When the classification weight of the information attribute is greater than the weight threshold w, the information attribute and content feature vector corresponding to the classification weight are collected and recorded in the key data list of the target file.

[0019] An information classification management system based on data analysis, the system comprising: Feature data sequence management module, reference data sequence management module, reference difference calculation module, feature information attribute management module and key data list management module; The characteristic data sequence management module is used to extract the information attributes and content vectors of the non-structured data respectively to obtain the characteristic data sequence, wherein the characteristic data sequence management module includes: an information attribute management unit, a content feature management unit and a characteristic data sequence management unit, wherein the information attribute management unit is used to manage the information attribute set of the non-structured data, the content feature management unit is used to manage the content vector set of the non-structured data, and the characteristic data sequence management unit is used to manage the characteristic data sequence of the non-structured data; The reference data sequence management module is used to manage the reference data sequence of the classification category, wherein the reference data sequence management module includes: a total attribute management unit, an attribute weight calculation unit, a reference vector calculation unit and a data collection unit, wherein the total attribute management unit is used to obtain the total number of attributes in the target category, the attribute weight calculation unit is used to calculate the attribute weight of the information attribute, the reference vector calculation unit is used to calculate the reference vector of the information attribute, and the data collection unit is used to collect the information attribute and the reference vector to obtain the reference data sequence; The reference difference calculation module is used to calculate the reference difference, wherein the reference difference calculation module includes: a vector group collection unit and a difference calculation unit, wherein the vector group collection unit is used to collect the first vector group and the second vector group, and the difference calculation unit is used to calculate the difference of the vector groups; The characteristic information attribute management module is used to collect information attribute subsets that are smaller than the reference difference, wherein the characteristic information attribute management module includes: a subset management unit and an attribute collection unit, wherein the subset management unit is used to obtain a subset in the vector group and construct a comparison attribute set, and the attribute collection unit is used to collect the characteristic attribute set to obtain information attributes with characteristic information attributes; Among them, the key data list management module is used to manage the key data list of unstructured data, wherein the key data list management module includes: a feature information statistics unit, a classification weight calculation unit and a key data collection unit, wherein the feature information statistics unit is used to obtain the number of times the information attribute is recorded as a feature information attribute, the classification weight calculation unit is used to calculate the classification weight of the information attribute, and the key data collection unit is used to collect information attributes and content feature vectors to obtain the key data list of the target file.

[0020] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. An information classification management method based on data analysis, characterized in that: The method comprises the following steps: Step S100: classifying the information of the non-structured data file, classifying and extracting the features of each type of information in the non-structured data file, and converting a non-structured data file into a corresponding feature data sequence; Step S200: Collect historical classification records of non-structured data, obtain characteristic data sequences of all non-structured data files in a classification category, and establish a reference data sequence of the classification category according to the numerical characteristics of the characteristic data sequence; Step S300: obtaining a non-structured data file from the classification category as a target file, taking a characteristic data sequence of the target file as a target data sequence, and recording a difference between the target data sequence and a reference data sequence as a reference difference; Step S400: Calculate the difference between the same information attribute subsets in the target data sequence and the reference data sequence, collect the information attribute subsets that are smaller than the reference difference, and record them as characteristic information attributes; Step S500: Obtain classification records including the target file from historical classification records, calculate the classification weight of each information attribute in the target file according to the frequency of each information attribute being recorded as a characteristic information attribute, and output a key data list of the target file.

2. The information classification management method based on data analysis according to claim 1 is characterized by: Step S100 includes: Step S101: obtaining i types of information attributes in a target file, and forming the information attributes of the target file into an information attribute set, wherein DA: a1, a2, a3, ..., ai, respectively represent the first, second, third, ..., and i-th information attributes of the target file; Step S102: Obtain the file content corresponding to the information attribute set DA, extract the feature vector of the file content to obtain the content feature vector, collect all the content feature vectors of the target file, and obtain the content vector set DB, DB: b1, b2, b3, ... bi, representing the content feature vectors corresponding to the first, second, third, ... and i-th information attributes of the target file respectively; Step S103: The corresponding information attributes and file contents are grouped into information groups, and all information groups of the target file are collected to obtain a characteristic data sequence of the target file, wherein the i-th information group in the characteristic data sequence is (ai, bi).

3. The information classification management method based on data analysis according to claim 2 is characterized in that: Step S200 includes: Step S201: record a classification category including a target file in a certain historical classification record as a target category, obtain n non-structured data files in the target category, and obtain a feature data sequence corresponding to each non-structured data file; Step S202: extracting information attribute sets of each unstructured data file respectively, calculating the intersection of the information attribute sets to obtain a classification attribute set, and obtaining the total number of information attributes in the classification attribute set, which is recorded as q; Step S203: Obtain the number r of all information attributes included in the e-th unstructured data in the target category e , calculate the total number of attributes Q in the target category, ; Step 204: In the target category, the number of occurrences of the same information attribute in all feature data sequences is traversed, where the number of occurrences of the jth information attribute in the classification attribute set is recorded as att j , calculate the attribute weight α of the j-th information attribute j , α j =att j / Q; Step S205: Obtain m content feature vectors of the j-th information attribute in the classification category, and calculate the reference vector ref of the j-th information attribute j , , where k f represents the fth content feature vector among the m content feature vectors; Step S206: Collect all information attributes in the target category and the reference vectors corresponding to each information attribute, and group the information attributes and the corresponding reference vectors into data groups to form a reference data sequence, wherein the g-th data group in the reference data sequence is (tri g , equal g ), tri g represents the information attribute in the g-th data group, g Indicates tri g The corresponding reference vector.

4. The information classification management method based on data analysis according to claim 3 is characterized by: Step S300 includes: Step S301: Obtain the classification attribute set P0, calculate the reference set P1, P1=P0∩DA, and obtain the reference vector corresponding to each information attribute in P1 from the reference data sequence; Step S302: Grouping the content feature vectors in the content vector set DB into a first vector group X of i dimensions i 1. The reference vectors corresponding to each information attribute in P1 are combined into a second vector group X of i dimensions i 2. Calculate the distance d0 between the first vector group and the second vector group, and record the reference distance as the difference between the target data sequence and the reference data sequence as the reference difference.

5. The information classification management method based on data analysis according to claim 4 is characterized in that: Step S400 includes: Step S401: Obtain h content feature vectors from the first vector group to form an h-dimensional vector group X h 1. Collect the information attributes corresponding to the h content feature vectors to obtain a comparison attribute set U; Step S402: Obtain reference vectors corresponding to each information attribute in the comparison attribute set from the second vector group, and collect the reference vectors to form an h-dimensional vector group X h 2, Calculate the vector group X h 1 and X h 2 The distance d h ; Step S403: When d h When <d0, the comparison attribute set U is recorded as a characteristic attribute set, the content characteristic vectors in the first vector group are traversed to obtain several characteristic attribute sets, the intersection of the several characteristic attribute sets is calculated, and all information attributes included in the intersection are recorded as characteristic information attributes.

6. The information classification management method based on data analysis according to claim 5 is characterized in that: Step S500 includes: Step S501: collect t times of historical classification records of the target file, collect the number of times each information attribute in the target file is recorded as a characteristic information attribute, where the number of times the i-th information attribute of the target file is recorded as a characteristic information attribute is mes i ; Step S502: Calculate the classification weight corresponding to each information attribute of the target file, where the classification weight of the i-th information attribute γi, γi=mes i / t; Step S503: setting a weight threshold w. When the classification weight of the information attribute is greater than the weight threshold w, the information attribute and content feature vector corresponding to the classification weight are collected and recorded in the key data list of the target file.

7. An information classification management system based on data analysis, used to execute an information classification management method based on data analysis according to any one of claims 1 to 6, characterized in that: The system includes: A feature data sequence management module, a reference data sequence management module, a reference difference calculation module, a feature information attribute management module and a key data list management module, wherein the feature data sequence management module is used to extract the information attributes and content vectors of the non-structured data respectively and collect them to obtain the feature data sequence, the reference data sequence management module is used to manage the reference data sequence of the classification category, the reference difference calculation module is used to calculate the reference difference, the feature information attribute management module is used to collect the information attribute subset that is smaller than the reference difference, and the key data list management module is used to manage the key data list of the non-structured data.

8. The information classification management system based on data analysis according to claim 7 is characterized by: The feature data sequence management module includes: an information attribute management unit, a content feature management unit and a feature data sequence management unit, wherein the information attribute management unit is used to manage the information attribute set of the non-structured data, the content feature management unit is used to manage the content vector set of the non-structured data, and the feature data sequence management unit is used to manage the feature data sequence of the non-structured data; The reference data sequence management module includes: a total attribute management unit, an attribute weight calculation unit, a reference vector calculation unit and a data collection unit, wherein the total attribute management unit is used to obtain the total number of attributes in the target category, the attribute weight calculation unit is used to calculate the attribute weight of the information attribute, the reference vector calculation unit is used to calculate the reference vector of the information attribute, and the data collection unit is used to collect the information attributes and the reference vector to obtain the reference data sequence.

9. The information classification management system based on data analysis according to claim 7, characterized in that: The reference difference calculation module includes: a vector group collection unit and a difference calculation unit, wherein the vector group collection unit is used to collect the first vector group and the second vector group, and the difference calculation unit is used to calculate the difference between the vector groups; The characteristic information attribute management module includes: a subset management unit and an attribute collection unit, wherein the subset management unit is used to obtain a subset in the vector group and construct a comparison attribute set, and the attribute collection unit is used to collect the characteristic attribute set to obtain information attributes of the existing characteristic information attributes.

10. The information classification management system based on data analysis according to claim 7, characterized in that: The key data list management module includes: a feature information statistics unit, a classification weight calculation unit and a key data collection unit, wherein the feature information statistics unit is used to obtain the number of times an information attribute is recorded as a feature information attribute, the classification weight calculation unit is used to calculate the classification weight of the information attribute, and the key data collection unit is used to collect information attributes and content feature vectors to obtain a key data list of the target file.

Images