Data processing method and device, computer device, and storage medium

Abstract

Embodiments of the present application disclose a data processing method and device, computer equipment and a storage medium, wherein the method comprises: obtaining a target data set, the target data set comprising target full name data and M corresponding short name data, M being a positive integer; identifying the data type of the M short name data, the data type comprising a suspicious type, the short name data of the suspicious type being short name data that cannot represent the semantic of the target full name data; if there is short name data of the suspicious type in the M short name data, performing data analysis on the target full name data to generate new short name data; and replacing the short name data of the suspicious type in the M short name data with the new short name data in the target data set, which can improve the effectiveness of data cleaning.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computer, in particular to a data processing method and device, computer equipment and storage medium. BACKGROUND

[0002] With the continuous development of computer technology, the amount of data stored in computer equipment is also increasing, such as when the computer equipment stores full name data and corresponding abbreviation data, a large amount of full-abbreviation data will be stored, and in order to facilitate subsequent business processing using full-abbreviation data, it is often necessary to first clean the full-abbreviation data stored in the computer equipment. The current method for cleaning data is denoising and deduplication, and based on the current method for cleaning data, the subsequent business cannot be effectively improved, so how to improve the effectiveness of data cleaning of full-abbreviation data has become a research hotspot. SUMMARY

[0003] The embodiment of the present application provides a data processing method, device, computer equipment and storage medium, which can improve the effectiveness of data cleaning.

[0004] In one aspect, the embodiment of the present application provides a data processing method, comprising:

[0005] obtaining a target data set, the target data set comprising target full name data and M abbreviation data corresponding to the target full name data, M being a positive integer;

[0006] identifying the data type of the M abbreviation data, the data type comprising a suspicious type, the abbreviation data of the suspicious type being abbreviation data that cannot represent the semantic of the target full name data;

[0007] if there is abbreviation data of the suspicious type in the M abbreviation data, performing data analysis on the target full name data to generate new abbreviation data;

[0008] replacing the abbreviation data of the suspicious type in the M abbreviation data with the new abbreviation data in the target data set.

[0009] In another aspect, the embodiment of the present application provides a data processing device, comprising:

[0010] an obtaining unit configured to obtain a target data set, the target data set comprising target full name data and M abbreviation data corresponding to the target full name data, M being a positive integer;

[0011] The recognition unit is configured to recognize data types of the M abbreviations, wherein the data types include a suspicious type, and the abbreviations of the suspicious type refer to abbreviations that cannot represent semantics of the target full name data.

[0012] The analysis unit is configured to perform data analysis on the target full name data to generate new abbreviations if there are abbreviations of the suspicious type in the M abbreviations.

[0013] The replacement unit is configured to replace the abbreviations of the suspicious type in the M abbreviations with the new abbreviations in the target data set.

[0014] In another aspect, an embodiment of the present application provides a computer device, comprising a processor, an input device, an output device and a memory, which are connected to each other, wherein the memory is configured to store a computer program supporting the computer device to execute the above method, the computer program comprises program instructions, and the processor is configured to invoke the program instructions to perform the following steps:

[0015] The target data set comprises target full name data and M abbreviations corresponding to the target full name data, and M is a positive integer.

[0016] The recognition unit is configured to recognize data types of the M abbreviations, wherein the data types include a suspicious type, and the abbreviations of the suspicious type refer to abbreviations that cannot represent semantics of the target full name data.

[0017] The analysis unit is configured to perform data analysis on the target full name data to generate new abbreviations if there are abbreviations of the suspicious type in the M abbreviations.

[0018] The replacement unit is configured to replace the abbreviations of the suspicious type in the M abbreviations with the new abbreviations in the target data set.

[0019] In another aspect, an embodiment of the present application provides a computer readable storage medium, wherein the computer readable storage medium stores program instructions, and the program instructions are executed by a processor to perform the data processing method of the first aspect.

[0020] In the embodiment of the present application, after the computer device determines the target full name data and the M short name data corresponding to the target full name data from the target data set, the computer device can identify the data type of each short name data in the M short name data, and filter out the short name data of suspicious type that cannot express the semantics of the target full name data from the M short name data. Based on the filtering of the short name data of suspicious type in the short name data, the subsequent data processing pressure of the computer device can be reduced, and the data processing efficiency of the computer device can be improved. After the computer device filters out the short name data of suspicious type, the target data is analyzed to generate new short name data, and the original short name data of suspicious type in the target data set is replaced by the newly generated short name data, so that the computer device can effectively improve the accuracy of the semantic expression of each short name data in the target data set to the corresponding full name data, which is beneficial to the accuracy when the target data set is used to perform downstream tasks. At the same time, based on the data analysis of the target full name data by the computer device and the replacement of the short name data of suspicious type by the newly generated short name data, the full name data and the short name data in the target data set can be enriched at the same time, and the data efficiency and sufficiency in the target data set are improved. BRIEF DESCRIPTION OF DRAWINGS

[0021] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0022] Figure 1 is a schematic diagram of a data search system provided by an embodiment of the present application;

[0023] Figure 2 is a schematic flow chart of a data processing method provided by an embodiment of the present application;

[0024] Figure 3 is a schematic diagram of data analysis of full name data provided by an embodiment of the present application;

[0025] Figure 4 is a schematic flow chart of a data processing method provided by an embodiment of the present application;

[0026] Figure 5 is a schematic diagram of a crawler verification provided by an embodiment of the present application;

[0027] Figure 6 is a schematic diagram of generating new short name data provided by an embodiment of the present application;

[0028] Figure 7 is a schematic diagram of a data processing method provided by an embodiment of the present application;

[0029] Figure 8 is a schematic block diagram of a data processing apparatus provided by an embodiment of the present application;

[0030] Figure 9 is a schematic block diagram of a computer device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0031] The embodiment of the present application provides a data processing method, which can enable the computer device to, when determining that there is suspicious type of abbreviated data in M abbreviated data of target full name data when performing data cleaning on a target data set, generate new abbreviated data based on the target full name data, so as to replace the suspicious type of abbreviated data corresponding to the target full name data by using the newly generated abbreviated data, so that the abbreviated data of the target full name data obtained by replacement is all abbreviated data capable of representing the semantics of the target full name data, so that the computer device can not only delete the abbreviated data in the target data set which cannot express the semantics of the full name data when performing data cleaning on the target data set, but also can supplement the abbreviated data in the target data set, so as to improve the effectiveness of data cleaning on the target data set, and significantly improve the number of abbreviated data in the target data set. In one embodiment, the target data set is any set including at least one abbreviated data and at least one full name data, the data types of the abbreviated data and the full name data included in the target data set can be any type, and any two full name data (or abbreviated data) in the target data set can be data of the same type or data of different types, and the embodiment of the present application does not limit the target full name data included in the target data set, which can be any full name data in the target data set. One full name data in the target data set can correspond to one or more abbreviated data, and one abbreviated data can also correspond to one or more full name data. It can be understood that the full name data is used to describe the official name of organizations and groups, and the abbreviated data is part of the entity word extracted from the full name data, which is used to summarize the full name data. The full name data can be, for example, "XXX Limited Liability Company", and the corresponding abbreviated data can be "XXX" or "XXX Company", etc. The original full name data in the target data set and the corresponding abbreviated data mentioned in the embodiment of the present application can be obtained from historical experience, and the embodiment of the present application does not limit the way of determining the original full name data in the target data set and the corresponding abbreviated data.

[0032] In an embodiment, after obtaining the target data set, the computer device can first perform data cleaning on the target data set. The data cleaning performed by the computer device on the abbreviation data and the full name data in the target data set includes denoising and deduplication, etc. Specifically, the computer device can remove illegal characters and characters that do not meet the length constraint, etc. in the target data set, and can also delete noise characters in the target data set. After the computer device performs data cleaning on the target data set, it can further combine expert experience to perform credibility evaluation on the full name data and the abbreviation data in the target data set. The full-abbreviation data group with low credibility is filtered out from the target data set through the label obtained by evaluation. Then, the computer device can perform crawler verification on the full-abbreviation data group with low credibility, so as to output credible abbreviations and non-credible abbreviations. After the computer device determines the credible abbreviations and the non-credible abbreviations from the target data set, it can perform text analysis on the full name data in the target data set, and generate new abbreviation data according to the result of text analysis. Then, the non-credible abbreviation data can be removed, and the corresponding credible new abbreviation data can be replaced, supplemented, etc. so as to obtain a new target data set, which can effectively increase the number of full-abbreviation data in the target data set and improve the accuracy of the abbreviation data in the target data set. In an embodiment, when performing text analysis on the full name data, the computer device can use a named entity recognition (NER) algorithm or other entity word recognition algorithm for processing.

[0033] After the computer device updates the full name data and the corresponding abbreviation data in the target data set to obtain a new target data set, the computer device can recommend search information based on the obtained new target data set. Specifically, the computer device can recommend search information based on the new target data set as follows: Figure 1As shown, after obtaining a new target dataset, computer device 10 can retrieve search information from terminal device 11. After obtaining the search information from terminal device 11, computer device 10 can search for full name information and / or abbreviation information matching the search information in the new target dataset. If computer device 10 determines that it has found abbreviation information matching the search information in the new target dataset, it can retrieve content data related to the abbreviation information and content data related to the corresponding full name information and feed it back to terminal device 11. The feedback content data can then be displayed on terminal device 11. Similarly, if computer device 10 finds full name data matching the search information in the new target dataset, it will retrieve corresponding content data based on the full name data and the corresponding abbreviation data as feedback data and feed it back to terminal device 11 for display. Feedback based on the new target dataset effectively improves the accuracy of the feedback content data, thereby increasing user satisfaction during the data search process. In one embodiment, the computer device may search for full-name data (or abbreviated data) that matches the semantics of the search information, or the computer device may extract keywords based on the search information to search for full-name data (or abbreviated data) that matches the keywords.

[0034] For a detailed explanation of the process of generating new abbreviation data corresponding to a full name in this target dataset, please refer to [link to documentation]. Figure 2 This is a schematic flowchart of a data processing method proposed in an embodiment of the present invention. This embodiment describes in detail the process of generating new abbreviation data for the full name data of the target dataset. The full name data of the target dataset can be any full name data in the target dataset, such as... Figure 2 As shown, the method may include:

[0035] S201, Obtain the target dataset. The target dataset includes the full name of the target and M abbreviations corresponding to the full name of the target, where M is a positive integer.

[0036] S202, Identify the data types of M abbreviation data, including suspicious types. Suspicious abbreviation data refers to abbreviation data that cannot represent the semantics of the target full name data.

[0037] The target data set obtained by the computer device includes one or more full name data, each of which corresponds to one or more abbreviated name data. Similarly, each abbreviated name data in the target data set also corresponds to one or more full name data. For example, the full name data is "XXX Limited Liability Company", and the corresponding abbreviated name data includes "XXX Company" or "XXX". The abbreviated name data "XXX" corresponds to the full name data "XXX Limited Liability Company" and "XXX Limited Liability Company-Shenzhen Branch". It can be understood that although one or more full name data and the corresponding abbreviated name relationship are recorded in the target data set, the computer device may make business errors when using the target data set for business processing due to the many-to-many relationship between the abbreviated name data and the full name data in the target data set. Therefore, the computer device needs to adjust the abbreviated name data in the target data set to obtain a new target data set to ensure the accuracy of business processing based on the new target data set.

[0038] In one embodiment, when the computer device adjusts the abbreviated name data in the target data set, the computer device can first perform data cleaning on the full name data and the abbreviated name data in the target data set. Specifically, the computer device can perform data deduplication, data cleaning, and noise removal on the full name data (or abbreviated name data) when performing data cleaning on the full name data and the abbreviated name data in the target data set. The data cleaning includes removing extra spliced data or extra English characters, and the noise removal includes removing illegal characters and data with a length that does not meet the constraint in the full name data (and abbreviated name data). Since data with a length that does not meet the constraint has no specific business semantics in most cases, the data with a length that does not meet the constraint can be deleted in the data cleaning stage. The constraint may be, for example, 2 characters or 3 characters. After the computer device performs data cleaning on the target data set, any full name data in the target data set can be selected as target full name data, and M abbreviated name data corresponding to the target full name data can be determined from the target data set.

[0039] After obtaining the target full name data and the corresponding M nickname data, the computer device can identify the data type of the M nickname data to determine the suspicious type of nickname data that cannot express the semantics of the target full name data from the M nickname data. In a specific implementation, when identifying the data type of the M nickname data, the computer device can perform semantic analysis on each of the M nickname data to determine the semantics of each nickname data. In addition, the computer device can also perform semantic analysis on the target full name data to determine the semantics of the target full name data. Then, the computer device can further perform semantic matching between the semantics of each nickname data and the semantics of the target full name data, and according to the result of semantic matching, find out the nickname data that cannot express the semantics of the target full name data (or cannot fully express the semantics of the target full name data, or has semantic ambiguity) from the M nickname data, wherein the found nickname data is the suspicious type of nickname data.

[0040] In one embodiment, after obtaining the target full name data and the corresponding M nickname data, the computer device can further perform further trust evaluation on the M nickname data to delete the untrusted nickname data in the M nickname data. When performing trust evaluation on the M nickname data, the computer device can perform trust evaluation on each of the M nickname data based on the data attribute of each of the M nickname data, such as by judging the data length of each nickname data. Then, after performing trust evaluation on each of the M nickname data, the computer device can perform association detection on each of the trust evaluation passed nickname data and the target full name data, and further determine whether each of the detected data can express the semantics of the target full name data based on the detection result of the association, so as to determine the data type of each of the nickname data. After the computer device determines the data type of each of the nickname data corresponding to the target full name data, the computer device can adjust the nickname data of the corresponding suspicious type in the target data set, so that the nickname data corresponding to the target full name data in the target data set can express the semantics of the target full name data.

[0041] S203, if there is nickname data of suspicious type in the M nickname data, the target full name data is parsed to generate new nickname data.

[0042] In one embodiment, after the computer device determines the data type of each of the M nickname data corresponding to the target full name data, the computer device can further perform data parsing processing on the target full name data, so that the computer device can generate new nickname data of the target full name data based on the parsing result of the target full name data. In one embodiment, when the computer device performs data parsing processing on the target full name data and generates corresponding new nickname data, the computer device can determine the semantic analysis result of the target full name data by performing semantic analysis on the target full name data to generate new nickname data. For example, if the target full name data is "Shanghai XX E-commerce Co., Ltd.", and the computer device determines that there is nickname data of suspicious type in the M nickname data corresponding to "Shanghai XX E-commerce Co., Ltd.", the computer device can perform semantic analysis on "Shanghai XX E-commerce Co., Ltd.", and if the generated semantic analysis result is "Shanghai XX Company", the semantic analysis result "Shanghai XX Company" can be used as the new nickname data corresponding to the target full name data.

[0043] In another implementation, when the computer device performs data parsing on the target full name data and generates corresponding new nickname data, the computer device can further call a sequence labeling model to parse the target full name data to determine the semantics of each entity word in the target full name data and the role expressed in the target full name data. Then, based on the parsing of the sequence labeling model on the target full name data, the computer device can extract one or more entity words from the target full name data for combination to obtain new nickname data corresponding to the target full name data. In one embodiment, the model structure of the sequence labeling model can be a bi-lstm+crf double-layer model structure, or a bert+lstm+crf three-layer model structure. When the computer device performs parsing on the target full name data and extracts one or more entity words from the target full name data, the computer device can extract and combine the entity words based on prior knowledge, or extract and combine the entity words through model training. Similarly, if the computer device determines that the target full name data is "Shanghai XX E-commerce Co., Ltd.", based on the calling of the sequence labeling model by the computer device, the computer device can determine the role corresponding to each entity word, such as Figure 3As shown, the entity word "Shanghai" corresponds to the role of a place name, the entity word "XX" corresponds to the role of a keyword, the entity word "e-commerce" corresponds to the role of an industry, and the entity word "co-operation company" corresponds to the role of general information. Based on the analysis of the target full name data, the computer device can extract one or more entity words from the target full name data to form new abbreviation data corresponding to the target full name data, based on the role corresponding to each entity word. The new abbreviation data obtained may be "Shanghai XX Company", "XX Co-operation Company", etc.

[0044] After the computer device parses the full name data of the target and generates new abbreviation data, the computer device can use the new abbreviation data to update the target dataset, that is, the computer device can then proceed to step S204.

[0045] S204, In the target dataset, replace the abbreviation data with a suspicious data type among the M abbreviation data with new abbreviation data.

[0046] When a computer device updates the target dataset using new abbreviation data, the computer device can replace the abbreviation data of the corresponding suspicious type with the new abbreviation data. In one embodiment, replacing the abbreviation data of the original suspicious type with the new abbreviation data means: in the target dataset, canceling the data association between the abbreviation data of the suspicious type and the target full name data, adding the generated new abbreviation data to the target dataset, and associating the added new abbreviation data with the target full name data. It should be noted that the number of new abbreviation data generated by the computer device based on the data parsing of the target full name data is one or more, and the number of new abbreviation data generated and the number of abbreviation data of the suspicious type may be the same or different. That is, the number of new abbreviation data generated by the computer device based on the data parsing of the target full name data may be greater than the number of abbreviation data of the suspicious type, may be equal to the number of abbreviation data of the suspicious type, or may be less than the number of abbreviation data of the suspicious type. This embodiment of the invention does not limit this.

[0047] In the embodiment of the present application, after the computer device determines the target full name data and the M short name data corresponding to the target full name data from the target data set, the computer device can identify the data type of each short name data in the M short name data, filter out the short name data of suspicious type which cannot express the semantic of the target full name data from the M short name data. Based on the filtering of the short name data of suspicious type in the short name data, the subsequent data processing pressure of the computer device can be reduced, and the data processing efficiency of the computer device can be improved. After the computer device filters out the short name data of suspicious type, the target data can be parsed to generate new short name data, and the original short name data of suspicious type in the target data set is replaced by the newly generated short name data, so that the computer device can effectively improve the accuracy of the semantic expression of each short name data in the target data set to the corresponding full name data, which is beneficial to the accuracy when the target data set is used to perform downstream tasks. At the same time, based on the data parsing of the target full name data by the computer device and the replacement of the short name data of suspicious type by the newly generated short name data, the full name data and the short name data in the target data set can be enriched at the same time, which improves the data efficiency and sufficiency in the target data set.

[0048] Please refer to Figure 4 is a schematic flow chart of a data processing method according to an embodiment of the present application, as shown in Figure 4 The method can include the following steps:

[0049] S401, obtaining a target data set, the target data set including target full name data and M short name data corresponding to the target full name data, M being a positive integer.

[0050] S402, identifying the data type of the M short name data, the data type including a suspicious type, the short name data of suspicious type being the short name data which cannot express the semantic of the target full name data.

[0051] In an embodiment, the target data set obtained by the computer device comprises one or more full name data and the associated short name data of each full name data, and the number of short name data associated with each full name data in the target data set can be one or more, and the number of full name data associated with each short name data can also be one or more. If the target full name data and M short name data associated with the target full name data are included in the target data set obtained by the computer device, and after the computer device obtains the target full name data and the corresponding M short name data, the computer device can identify the data type of each short name data in the M short name data to determine the data type of each short name data. When the computer and the device perform data identification, the computer device can first obtain the data attribute of each short name data in the M short name data, the data attribute of the target full name data, and the data association between the target full name data and each short name data. Then, the computer device can perform a trust evaluation on each short name data according to the data attribute of each short name data, the data attribute of the target full name data, and the data association, and determine the data type of the M short name data according to the trust evaluation result.

[0052] The data attribute of the short name data (or the target full name data) is data used for qualitative description and / or quantitative description of the short name data (or the target full name data), such as data length, data content included in the data, etc. Therefore, when the computer device performs a trust evaluation on each short name data based on the data attribute of the short name data, the data attribute of the target full name data, and the data association, the computer device can evaluate a label for the corresponding short name data based on one or more evaluation methods. The evaluation label is used to indicate whether the corresponding short name data is trustworthy based on the trust evaluation. In a specific implementation, the computer device can add a suspicious label to any short name data when it is determined that the data length (such as 2 or 3, etc.) of any short name data corresponding to the data attribute of the short name data is less than a length threshold. The suspicious label is an evaluation label added by the computer device to the any short name data based on the trust evaluation of the any short name data, and the suspicious label is used to indicate that the computer device considers the any short name data to be temporarily untrustworthy after performing the trust evaluation on the any short name data. In addition, the computer device can also determine whether there is an intersection between the any short name data and the target full name data according to the data attribute of the any short name data, the data attribute of the target full name data, and the data association when performing the trust evaluation on the short name data, and add a trust label to the any short name data when it is determined that there is an intersection. It can be understood that the trust label is used to indicate that the any short name data is temporarily trustworthy after the trust evaluation.

[0053] In one embodiment, when the computer device adds the corresponding evaluation label to the abbreviation data based on the result of the trusted evaluation of the abbreviation data, the computer device can further determine whether any abbreviation data and the target full name data are one-to-one corresponding, and add a trusted label to the any abbreviation data when it is determined to be one-to-one corresponding, and add a suspicious label to the any abbreviation data when it is not one-to-one corresponding. In addition, in the embodiments of the present application, the manner and number of evaluation means used by the computer device to perform the trusted evaluation of the abbreviation data are not limited, that is, the computer device can select one or more of the above-mentioned three ways of trusted evaluation of the abbreviation data to perform the trusted evaluation of the abbreviation data. When there are multiple ways of trusted evaluation of the abbreviation data in the computer device, the multiple evaluation methods can be sorted, and then the corresponding evaluation method is used to perform the trusted evaluation of a certain abbreviation data to obtain the evaluation result of the certain abbreviation data. For example, if the computer device determines that the abbreviation data is "abc" and the target full name data is "Shanghai XX E-commerce Co., Ltd.", since there is no intersection between the abbreviation data "abc" and the target full name data "Shanghai XX E-commerce Co., Ltd.", it is determined that the abbreviation data "abc" is not trusted, and a suspicious label is added to the abbreviation data.

[0054] After the computer device determines the data type of the M abbreviation data according to the trusted evaluation result, it can filter out the abbreviation data with a trusted label from the M abbreviation data according to the trusted evaluation result, and the data type of the abbreviation data with a trusted label is a trusted type, which means that the abbreviation data can represent the semantics of the target full name data. In addition, the computer device can also determine the abbreviation data with a suspicious label from the M abbreviation data. After the computer device determines the abbreviation data with a suspicious label, it can perform semantic analysis on the abbreviation data with a suspicious label and the target full name data to obtain a semantic analysis result. Then, the computer device can determine the semantic association between the semantics of the abbreviation data with a suspicious label and the semantics of the target full name data according to the semantic analysis result, so as to determine the data type of the abbreviation data with a suspicious label according to the semantic association. That is, after the computer device performs the trusted evaluation of the abbreviation data, it can further analyze the abbreviation data that is evaluated as suspicious based on the result of the trusted evaluation, so as to determine the data type of the abbreviation data based on the result of the secondary analysis, thereby effectively improving the reliability of the data type of each abbreviation data determined by the computer device.

[0055] When the computer device performs semantic analysis on the short name data with the suspicious label added and the target full name data, the computer device can first take the short name data with the suspicious label added as a first crawler keyword and take the target full name data as a second crawler keyword, so that the computer device can perform crawler search according to the first crawler keyword and the second crawler keyword to obtain a crawler search result, and take the crawler search result as the result of the semantic analysis on the short name data with the suspicious label added and the target full name data. Then, when the computer device determines the semantic association between the semantic of the short name data with the suspicious label added and the semantic of the target full name data according to the semantic analysis result, the computer device can determine that the semantic of the short name data with the suspicious label added and the semantic of the target full name data are associated when the crawler search result indicates that the first crawler keyword and the second crawler keyword appear together in the search result, and the computer device can determine that the semantic of the short name data with the suspicious label added and the semantic of the target full name data are not associated when the computer device determines that the crawler search result indicates that the first crawler keyword and the second crawler keyword do not appear together in the search result. For example, if the computer device determines that the target full name data is “Shanghai XX E-commerce Co., Ltd.” and the short name data with the suspicious label added includes “abc” based on the above-mentioned trusted verification, the computer device can perform crawler search by taking “Shanghai XX E-commerce Co., Ltd.” as the second crawler keyword and taking “abc” as the first crawler keyword. If the search result obtained by performing crawler search based on the two crawler keywords is as shown in Table 1, the computer device can determine, based on the search result, that “abc” as the first crawler keyword and “Shanghai XX E-commerce Co., Ltd.” as the second crawler keyword do not appear together, and then mark the short name data “abc” as an untrusted state. It can be understood that the data type corresponding to the short name data “abc” marked as the untrusted state is the suspicious type. Figure 5

[0056] Based on the above-mentioned multiple analysis of the data type of the short name data, the data type corresponding to each short name data is determined, which can effectively improve the accuracy in determining the data type of each short name data. After the computer device determines the data type of each short name data, the computer device can perform data analysis on the target full name data if the target full name data includes the short name data of the suspicious type, so as to replace the short name data of the suspicious type, that is, to perform step S403.

[0057] S403, if there is short name data of the suspicious type in the M short name data, performing data analysis on the target full name data to generate new short name data.

[0058] ​S404, replace the nickname data of the suspicious type in the M nickname data with new nickname data in the target dataset.

[0059] In one embodiment, the target full name data includes one or more entity words, and the computer device, when performing data parsing on the target full name data to generate new nickname data, can perform data parsing on the target full name data to determine the entity role of each entity word in the target full name data, and then can combine any one or more entity words according to the entity role of each entity word, and take the combined entity words as new nickname data. The computer device can use a sequence labeling model to parse the target full name data, and the sequence labeling model is a trained model for identifying the role of entity words. The parsing process of the computer device using the sequence labeling model on the target full name data can be as shown in Figure 3 After determining the entity role of each entity word in the target full name data, the computer device can combine any one or more entity words to obtain new nickname data, as shown in Figure 6 For example, the computer device can combine entity words according to the roles of the entity words and the dependency relationship between the roles to obtain new nickname data.

[0060] In one embodiment, if the target full name data is "Shanghai XX E-commerce Co., Ltd.", the new nickname data can be obtained by recombining the entity words Shanghai, XX, e-commerce, and limited company based on the dependency relationship between the corresponding roles of the four entity words. The new nickname data can be "Shanghai XX", "XX e-commerce", or "XX e-commerce", etc. After the computer device generates new nickname data, the corresponding nickname data of the suspicious type in the target dataset can be replaced with the new nickname data to improve the reliability of each nickname data in the target dataset.

[0061] S405, according to the data type of each nickname data, screen out the nickname data of the trusted type from the M nickname data.

[0062] S406, if the nickname data of the trusted type corresponds to at least two full name data, screen out the commonly used full name data and the rarely used full name data from the at least two full name data.

[0063] S407, set a first weight value for the commonly used full name data and a second weight value for the rarely used full name data, and the priority of the first weight value is higher than the priority of the second weight value.

[0064] In steps S405-S407, based on the data type of each abbreviation data determined by the computer device, the computer device can further filter out the abbreviation data of the trusted type from the M abbreviation data of the target full name data, and after filtering out the abbreviation data of the trusted type, if one abbreviation data of the trusted type corresponds to at least two full name data, the computer device can filter out the commonly used full name data and the rarely used full name data from the corresponding at least two full name data, wherein the commonly used full name data corresponding to the abbreviation data of the trusted type refers to: based on the daily expression needs of the user, the full name data referred to by the abbreviation data of the trusted type, such as based on the daily expression needs of the user, the abbreviation data "PC" is commonly used to refer to a computer rather than a mobile phone, so if the full name data corresponding to the abbreviation data in the target data set includes a computer and a mobile phone, then after the computer device determines that the abbreviation data "PC" is a trusted abbreviation data, the commonly used full name data determined by the computer device based on the abbreviation data is a computer, and the rarely used full name data is a mobile phone. Then, after the computer device filters out the commonly used full name data and the rarely used full name data based on the trusted abbreviation data, the computer device can set a first weight value for the commonly used full name data and a second weight value for the rarely used full name data, and the priority of the first weight value is higher than the priority of the second weight value. It can be understood that when one abbreviation data corresponds to multiple full name data, the computer device can maintain a commonly used data dictionary in the target data set by setting the weight value priority of the corresponding different full name data, and through the maintenance of the commonly used data dictionary, the one-to-many relationship between the full name data and the abbreviation data in the target data set can be effectively reduced, thereby improving the accuracy of performing downstream tasks using the target data set.

[0065] In one embodiment, the commonly used data dictionary maintained by the computer device in the target data set can be maintained based on whether the full name data corresponds to a head enterprise, wherein the head enterprise refers to an enterprise that has a leading role in the relevant field, i.e., if the computer device determines that the full name data corresponding to the abbreviation data of the trusted type includes the company name of company A and the company name of company B, and in the field of the Internet industry, if company A is a head enterprise in the field and company B is not a head enterprise in the field, then when the computer device filters out the commonly used full name data and the rarely used full name data corresponding to the abbreviation data of the trusted type, the computer device can take the company name of company A as the commonly used full name data and the company name of company B as the rarely used full name data.

[0066] The computer device can maintain a commonly used data dictionary in the target data set based on screening of commonly used full name data and uncommon full name data. The commonly used data dictionary can be applied to downstream tasks. For example, when the downstream task is a search task, the computer device can obtain target search information, which includes reference abbreviation data of a trusted type. Then, data related to the reference abbreviation data and the commonly used full name data corresponding to the reference abbreviation data can be obtained as search result data. The search result data can be displayed on the terminal device. The search feedback based on commonly used full name data can make the search feedback results more in line with user needs, effectively reduce the data processing pressure of the computer device, and improve the data processing efficiency of the computer device. In an embodiment, after determining the data type of the abbreviation data, the computer device can screen the abbreviation data of a trusted type from the M abbreviation data according to the data type of each abbreviation data in the M abbreviation data, and determine one or more other full name data corresponding to the abbreviation data of a trusted type. Then, the target full name data and the one or more other full name data can be analyzed to generate recommended abbreviation data, and the target full name data and the one or more other full name data can be added with the recommended abbreviation data. Adding the recommended abbreviation data to the full name data can significantly increase the amount of data corresponding to the full name data and the abbreviation data in the target data set.

[0067] Based on the above process of processing any target full name data in the target data set, the following describes the case where the computer device processes each full name data in the target data set. Figure 7 When the computer device processes each full name data in the target data set, the specific process of processing any full name data in the target data set by the computer device can be referred to the description of the above embodiments. When the computer device processes the target data set, the computer device mainly needs the abbreviation screening module, the verification module, and the full name data analysis model. The computer device can first use the abbreviation screening module to clean, denoise, and remove duplicates of the full name data in the target data set. Then, the computer device can perform trust evaluation on the processed abbreviation data to obtain corresponding evaluation labels. Based on the trust evaluation, the abbreviation screening module can send the abbreviation data with a suspicious label to the verification module. After obtaining the suspicious abbreviation data, the verification module can perform web crawler processing based on the suspicious abbreviation data to obtain crawler verification results. Then, the data type of the abbreviation data is determined based on the crawler verification results. It should be noted that the abbreviation data with a trusted label determined based on the trust evaluation can be directly regarded as (temporary) trusted abbreviation data in the verification module.

[0068] After the verification module determines credible and acceptable abbreviation data based on the crawler verification results and the input of the credibility assessment, for any suspicious abbreviation data, the verification module can input the corresponding full name data into the full name data parsing module. This allows the full name data parsing module to generate new abbreviation data corresponding to the full name data of the suspicious abbreviation data, and then replace the suspicious abbreviation data with the new abbreviation data. For credible abbreviation data, the computer equipment can generate other recommended abbreviation data and maintain a commonly used data dictionary, thereby effectively adjusting the target dataset.

[0069] In this embodiment of the invention, after determining the full name data of the target from the target dataset, the computer device can identify the data type of each abbreviation corresponding to the full name data to determine the data type of each abbreviation. Then, when the abbreviation data corresponding to the full name data includes suspicious abbreviation data, the computer device can parse the full name data and generate new abbreviation data based on the parsing results. This new abbreviation data is then used to replace the suspicious abbreviation data in the target dataset. Furthermore, the computer device can filter out trustworthy abbreviation data and, based on the full name data corresponding to the trustworthy abbreviation data, filter out commonly used and uncommon full name data. By setting weight values, different priorities can be assigned to commonly used and uncommon full name data. By distinguishing and maintaining the commonly used and uncommon full name data, the reliability of the computer device in subsequent tasks using the adjusted target dataset can be effectively improved.

[0070] Based on the description of the above data processing method embodiments, this invention also proposes a data processing apparatus. This data processing apparatus can be a computer program (including program code) running on the aforementioned computer device, wherein the computer device can be a terminal device or a server device. This data processing apparatus can be used to perform operations such as... Figure 2 and Figure 4 For the data processing method described above, please refer to [link / reference]. Figure 8 The data processing device includes: an acquisition unit 801, an identification unit 802, a parsing unit 803, and a replacement unit 804.

[0071] The acquisition unit 801 is used to acquire a target dataset, which includes the full name data of the target and M abbreviation data corresponding to the full name data of the target, where M is a positive integer.

[0072] The identification unit 802 is used to identify the data types of the M abbreviation data, wherein the data types include suspicious types, and the abbreviation data of the suspicious types refers to abbreviation data that cannot represent the semantics of the full name of the target data;

[0073] The parsing unit 803 is configured to perform data parsing on the target full name data to generate new nickname data if the M nickname data includes nickname data of a suspicious type.

[0074] The replacing unit 804 is configured to replace the nickname data of the suspicious type in the M nickname data with the new nickname data in the target data set.

[0075] In an embodiment, the identification unit 802 is specifically configured to:

[0076] obtain a data attribute of each of the M nickname data, a data attribute of the target full name data, and a data correlation between the target full name data and each of the nickname data;

[0077] perform a trust evaluation on each of the nickname data according to the data attribute of each of the nickname data, the data attribute of the target full name data, and the data correlation, and determine a data type of the M nickname data according to the trust evaluation result.

[0078] In an embodiment, the identification unit 802 is specifically configured to:

[0079] add a suspicious label to any of the nickname data if it is determined that a data length of the any of the nickname data is less than a length threshold according to a data attribute corresponding to the any of the nickname data; or

[0080] add a trust label to the any of the nickname data if it is determined that the any of the nickname data and the target full name data have an intersection according to the data attribute corresponding to the any of the nickname data, the data attribute of the target full name data, and the data correlation.

[0081] In an embodiment, the data type further includes a trust type, and the nickname data of the trust type refers to nickname data capable of representing semantics of the target full name data; the identification unit 802 is specifically configured to:

[0082] filter out, from the M nickname data, nickname data to which the trust label is added according to the trust evaluation result, and the data type of the nickname data to which the trust label is added is the trust type;

[0083] determine, from the M nickname data, nickname data to which the suspicious label is added, and perform semantic analysis on the nickname data to which the suspicious label is added and the target full name data to obtain a semantic analysis result;

[0084] According to the semantic analysis result, determine semantic relevance between semantics of the suspiciously-labeled short name data and semantics of the target full name data, and determine a data type of the suspiciously-labeled short name data according to the semantic relevance.

[0085] In one embodiment, the identification unit 802 is specifically configured to:

[0086] The suspiciously-labeled short name data is taken as a first crawler keyword, and the target full name data is taken as a second crawler keyword.

[0087] According to the first crawler keyword and the second crawler keyword, crawler search is performed to obtain a crawler search result, and the crawler search result is taken as a result of semantic analysis on the suspiciously-labeled short name data and the target full name data.

[0088] In one embodiment, the identification unit 802 is specifically configured to:

[0089] If the crawler search result indicates that the first crawler keyword and the second crawler keyword coexist in the search result, it is determined that the semantics of the suspiciously-labeled short name data and the semantics of the target full name data are associated.

[0090] If the crawler search result indicates that the first crawler keyword and the second crawler keyword do not coexist in the search result, it is determined that the semantics of the suspiciously-labeled short name data and the semantics of the target full name data are irrelevant.

[0091] In one embodiment, the target full name data includes one or more entity words; the analysis unit 803 is specifically configured to:

[0092] Perform named entity recognition processing on the target full name data to determine an entity role of each entity word in the target full name data.

[0093] According to the entity role of each entity word, combine any one or more entity words, and take the combined entity word as new short name data.

[0094] In one embodiment, the data type further includes a trusted type, and the short name data of the trusted type refers to short name data capable of representing semantics of corresponding full name data; the apparatus further includes a screening unit 805 and a setting unit 806.

[0095] The screening unit 805 is configured to screen, according to the data type of each short name data, short name data of the trusted type from the M short name data.

[0096] The screening unit 805 is further configured to, if the short form data of the trusted type corresponds to at least two full form data, screen, from the at least two full form data, commonly used full form data and less commonly used full form data.

[0097] The setting unit 806 is configured to set a first weight value for the commonly used full form data and a second weight value for the less commonly used full form data, and the first weight value has a higher priority than the second weight value.

[0098] In an embodiment, the apparatus further includes a display unit 807.

[0099] The obtaining unit 801 is further configured to obtain target search information from a terminal device, the target search information including reference short form data of a trusted type.

[0100] The obtaining unit 801 is further configured to obtain data related to the reference short form data and commonly used full form data corresponding to the reference short form data as search result data.

[0101] The display unit 807 is configured to display the search result data in the terminal device.

[0102] In an embodiment, the data type further includes a trusted type, the short form data of the trusted type refers to short form data capable of representing semantics of corresponding full form data, and the apparatus further includes an adding unit 808.

[0103] The screening unit 805 is further configured to screen, from the M short form data, short form data of a trusted type according to a data type of each short form data in the M short form data, and determine one or more other full form data corresponding to the short form data of the trusted type.

[0104] The analyzing unit 803 is further configured to perform data analysis on the target full form data and the one or more other full form data.

[0105] The adding unit 808 is configured to generate recommended short form data and add the recommended short form data to the target full form data and the one or more other full form data.

[0106] In the embodiment of the present application, after the target full name data and the M short name data corresponding to the target full name data are determined from the target data set, the recognition unit 802 can recognize the data type of each short name data in the M short name data, to filter out the short name data of suspicious type which cannot express the semantics of the target full name data from the M short name data. Based on the filtering of the short name data of suspicious type in the short name data, the subsequent data processing pressure of the computer device can be reduced, and the data processing efficiency can be improved. After the short name data of suspicious type is filtered out, the analysis unit 803 can analyze the target data to generate new short name data, so that the replacement unit 804 can replace the original short name data of suspicious type in the target data set with the newly generated short name data, which can effectively improve the accuracy of the semantic expression of each short name data in the target data set to the corresponding full name data, and is beneficial to the accuracy when the target data set is used to perform downstream tasks. At the same time, based on the data analysis of the target full name data and the replacement of the short name data of suspicious type with the newly generated short name data, the full name data and the short name data in the target data set can be enriched at the same time, and the data efficiency and sufficiency in the target data set are improved.

[0107] Please refer to Figure 9 is a structural schematic block diagram of a computer device provided by an embodiment of the present application. As shown in the computer device in the embodiment can include one or more processors 901, one or more input devices 902, one or more output devices 903 and a memory 904. The above-mentioned processor 901, input device 902, output device 903 and memory 904 are connected through a bus 905. The memory 904 is used to store a computer program, the computer program includes program instructions, and the processor 901 is used to execute the program instructions stored in the memory 904. Figure 9

[0108] The memory 904 can include a volatile memory such as a random-access memory (RAM), and the memory 904 can also include a non-volatile memory such as a flash memory, a solid-state drive (SSD) and the like, and the memory 904 can also include a combination of the above-mentioned kinds of memories.

[0109] ​The processor 901 can be a central processing unit (CPU). The processor 901 can further include a hardware chip. The hardware chip can be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or the like. The PLD can be a field-programmable gate array (FPGA), a generic array logic (GAL), or the like. The processor 901 can also be a combination of the above structures.

[0110] In the embodiment of the present application, the memory 904 is configured to store a computer program, and the processor 901 is configured to execute program instructions stored in the memory 904 to implement the above-mentioned method. Figure 2 and Figure 4 corresponding steps of the method.

[0111] In one embodiment, the processor 901 is configured to execute the program instructions to perform the following steps:

[0112] obtain a target data set, the target data set including a target full name data and M short name data corresponding to the target full name data, M being a positive integer;

[0113] identify data types of the M short name data, the data types including a suspicious type, the short name data of the suspicious type being short name data that cannot represent the semantic of the target full name data;

[0114] if there is short name data of the suspicious type in the M short name data, perform data analysis on the target full name data to generate new short name data;

[0115] replace, in the target data set, the short name data of the suspicious type in the M short name data with the new short name data.

[0116] In one embodiment, the processor 901 is configured to execute the program instructions to perform the following steps:

[0117] obtain data attributes of each of the M short name data, a data attribute of the target full name data, and data correlation between the target full name data and each of the short name data;

[0118] According to the data attribute of each abbreviation data, the data attribute of the target full name data, and the data correlation, a trusted evaluation is performed on each abbreviation data, and a data type of the M abbreviation data is determined according to the trusted evaluation result.

[0119] In one embodiment, the processor 901 is configured to invoke the program instructions for performing:

[0120] If it is determined according to the data attribute corresponding to any abbreviation data that the data length of the any abbreviation data is less than a length threshold, a suspicious label is added to the any abbreviation data; or,

[0121] If it is determined according to the data attribute corresponding to the any abbreviation data, the data attribute of the target full name data, and the data correlation that the any abbreviation data and the target full name data have an intersection, a trusted label is added to the any abbreviation data.

[0122] In one embodiment, the data type further includes a trusted type, and the abbreviation data of the trusted type refers to abbreviation data capable of representing the semantics of the target full name data; the processor 901 is configured to invoke the program instructions for performing:

[0123] According to the trusted evaluation result, the abbreviation data with the trusted label is filtered out from the M abbreviation data, and the data type of the abbreviation data with the trusted label is the trusted type;

[0124] The abbreviation data with the suspicious label is determined from the M abbreviation data, and semantic analysis is performed on the abbreviation data with the suspicious label and the target full name data to obtain a semantic analysis result;

[0125] According to the semantic analysis result, a semantic correlation between the semantics of the abbreviation data with the suspicious label and the semantics of the target full name data is determined, and a data type of the abbreviation data with the suspicious label is determined according to the semantic correlation.

[0126] In one embodiment, the processor 901 is configured to invoke the program instructions for performing:

[0127] The abbreviation data with the suspicious label is taken as a first crawler keyword, and the target full name data is taken as a second crawler keyword;

[0128] Crawler search is performed according to the first crawler keyword and the second crawler keyword to obtain a crawler search result, and the crawler search result is taken as a result of semantic analysis on the abbreviation data with the suspicious label and the target full name data.

[0129] In one embodiment, the processor 901 is configured to invoke the program instructions for performing:

[0130] If the crawler search result indicates that the first crawler keyword and the second crawler keyword do not co-occur in the search result, it is determined that the semantics of the suspiciously-labeled acronym data and the semantics of the target full name data are irrelevant.

[0131] If the crawler search result indicates that the first crawler keyword and the second crawler keyword do not co-occur in the search result, it is determined that the semantics of the suspiciously-labeled acronym data and the semantics of the target full name data are irrelevant.

[0132] In one embodiment, the target full name data includes one or more entity words; the processor 901 is configured to invoke the program instructions for performing:

[0133] Performing named entity recognition processing on the target full name data to determine the entity role of each entity word in the target full name data;

[0134] According to the entity role of each entity word, combining any one or more entity words, and taking the combined entity word as new acronym data.

[0135] In one embodiment, the data type further includes a trusted type, and the acronym data of the trusted type refers to acronym data that can represent the semantics of the corresponding full name data; the processor 901 is configured to invoke the program instructions for performing:

[0136] According to the data type of each acronym data, filtering out the acronym data of the trusted type from the M acronym data;

[0137] If the acronym data of the trusted type corresponds to at least two full name data, filtering out commonly used full name data and rarely used full name data from the at least two full name data;

[0138] Setting a first weight value for the commonly used full name data and a second weight value for the rarely used full name data, and the priority of the first weight value is higher than the priority of the second weight value.

[0139] In one embodiment, the processor 901 is configured to invoke the program instructions for performing:

[0140] Obtaining target search information from a terminal device, the target search information including reference acronym data of a trusted type;

[0141] Obtaining data related to the reference acronym data and commonly used full name data corresponding to the reference acronym data as search result data;

[0142] displaying the search result data in the terminal device.

[0143] In one embodiment, the data type further includes a trusted type, and the short name data of the trusted type refers to short name data capable of representing the semantic of the corresponding full name data; the processor 901 is configured to invoke the program instruction for performing:

[0144] According to the data type of each of the M short name data, the short name data of the trusted type is filtered from the M short name data, and one or more other full name data corresponding to the short name data of the trusted type is determined;

[0145] The target full name data and the one or more other full name data are subjected to data analysis to generate recommended short name data, and the recommended short name data is added to the target full name data and the one or more other full name data.

[0146] The embodiment of the present application provides a computer program product or computer program, which comprises computer instructions stored in a computer readable storage medium. The processor of the computer device reads the computer instructions from the computer readable storage medium, and the processor executes the computer instructions, so that the computer device executes the above-mentioned method embodiment as shown in the Figure 2 or Figure 4 The computer readable storage medium can be a disc, an optical disc, a read-only memory (ROM) or a random access memory (RAM), etc.

[0147] The above only discloses partial embodiments of the present application, and of course cannot limit the scope of the present application. Those skilled in the art can understand that all or part of the above-mentioned embodiments can be implemented, and equivalent changes made according to the claims of the present application still fall within the scope of the present application.

Claims

1. A data processing method, characterized by, The method comprises the following steps: obtaining a target data set, wherein the target data set comprises target full name data and M pieces of short name data corresponding to the target full name data, and M is a positive integer; performing credibility evaluation on the M pieces of short name data, and identifying the data type of the M pieces of short name data according to the credibility evaluation result, wherein the data type comprises a suspicious type, the short name data of the suspicious type refers to short name data that cannot represent the semantic meaning of the target full name data, and the credibility evaluation result is used to add a credibility label or a suspicious label to the short name data, and when the data type of the M pieces of short name data is identified according to the credibility evaluation result, the identification is based on the semantic analysis result of the short name data to which the suspicious label is added and the target full name data; obtaining the semantic analysis result of the short name data to which the suspicious label is added and the target full name data in the following manner: taking the short name data to which the suspicious label is added as a first crawler keyword, taking the target full name data as a second crawler keyword, performing crawler search according to the first crawler keyword and the second crawler keyword to obtain a crawler search result, and taking the crawler search result as the semantic analysis result of the short name data to which the suspicious label is added and the target full name data; if there is short name data of the suspicious type in the M pieces of short name data, performing data analysis on the target full name data to generate new short name data; in the target data set, replacing the short name data of the suspicious type in the M pieces of short name data with the new short name data.

2. The method of claim 1, wherein, The identification of the data type of the M pieces of short name data comprises the following steps: obtaining the data attribute of each piece of short name data in the M pieces of short name data, the data attribute of the target full name data, and the data correlation between the target full name data and each piece of short name data; performing credibility evaluation on each piece of short name data according to the data attribute of each piece of short name data, the data attribute of the target full name data, and the data correlation, and determining the data type of the M pieces of short name data according to the credibility evaluation result.

3. The method of claim 2, wherein, The credibility evaluation on each piece of short name data comprises the following steps: if it is determined according to the data attribute of any piece of short name data that the data length of the any piece of short name data is less than a length threshold, adding a suspicious label to the any piece of short name data; or if it is determined according to the data attribute of the any piece of short name data, the data attribute of the target full name data, and the data correlation that the any piece of short name data and the target full name data have an intersection, adding a credibility label to the any piece of short name data.

4. The method of claim 2, wherein, The data type further comprises a credibility type, the short name data of the credibility type refers to short name data that can represent the semantic meaning of the target full name data, and the determination of the data type of the M pieces of short name data according to the credibility evaluation result comprises the following steps: according to the credibility evaluation result, screening short name data to which a credibility label is added from the M pieces of short name data, and the data type of the short name data to which the credibility label is added is the credibility type. determine suspiciously-labeled acronym data from the M pieces of acronym data, and perform semantic analysis on the suspiciously-labeled acronym data and the target full name data to obtain a semantic analysis result; determine semantic relevance between semantics of the suspiciously-labeled acronym data and semantics of the target full name data according to the semantic analysis result, and determine a data type of the suspiciously-labeled acronym data according to the semantic relevance.

5. The method of claim 1, wherein, The determining of the semantic relevance between the semantics of the suspiciously-labeled acronym data and the semantics of the target full name data according to the semantic analysis result comprises: if the crawler search result indicates that the first crawler keyword and the second crawler keyword co-occur in the search result, it is determined that the semantics of the suspiciously-labeled acronym data and the semantics of the target full name data are associated; if the crawler search result indicates that the first crawler keyword and the second crawler keyword do not co-occur in the search result, it is determined that the semantics of the suspiciously-labeled acronym data and the semantics of the target full name data are irrelevant.

6. The method of claim 1, wherein, The target full name data comprises one or more entity words. The data analysis on the target full name data to generate new acronym data comprises: performing named entity recognition processing on the target full name data to determine an entity role of each entity word in the target full name data; combining any one or more entity words according to the entity role of each entity word, and taking the combined entity words as new acronym data.

7. The method of claim 1, wherein, The data type further comprises a trusted type, and the acronym data of the trusted type refers to acronym data capable of representing semantics of corresponding full name data; the method further comprises: screening, from the M pieces of acronym data, acronym data of the trusted type according to the data type of each piece of acronym data; if the acronym data of the trusted type corresponds to at least two full name data, screening, from the at least two full name data, commonly-used full name data and uncommonly-used full name data; setting a first weight value for the commonly-used full name data and a second weight value for the uncommonly-used full name data, and the first weight value has a higher priority than the second weight value.

8. The method of claim 7, wherein, The method further comprises: obtaining target search information from a terminal device, the target search information comprising reference acronym data of the trusted type; obtaining data related to the reference acronym data and commonly-used full name data corresponding to the reference acronym data as search result data; displaying the search result data in the terminal device.

9. The method of claim 1, wherein, The data type further comprises a trusted type, and the acronym data of the trusted type refers to acronym data capable of representing semantics of corresponding full name data; the method further comprises: screening, from the M pieces of acronym data, acronym data of the trusted type according to the data type of each piece of acronym data in the M pieces of acronym data, and determining one or more other full name data corresponding to the acronym data of the trusted type. The target full name data and the one or more other full name data are subjected to data analysis, recommended abbreviation data is generated, and the recommended abbreviation data is added to the target full name data and the one or more other full name data.

10. A data processing apparatus, characterized by: The method comprises the following steps: An acquisition unit is configured to acquire a target data set, wherein the target data set comprises target full name data and M pieces of abbreviation data corresponding to the target full name data, and M is a positive integer. An identification unit is configured to perform trust evaluation on the M pieces of abbreviation data, and identify data types of the M pieces of abbreviation data according to the trust evaluation result, wherein the data types include a suspicious type, the abbreviation data of the suspicious type refers to abbreviation data that cannot represent the semantics of the target full name data, and the trust evaluation result is used to add a trust label or a suspicious label to the abbreviation data, and the identification of the data types of the M pieces of abbreviation data is based on the abbreviation data added with the suspicious label and a semantic analysis result of the target full name data. The manner of obtaining the abbreviation data added with the suspicious label and the semantic analysis result of the target full name data comprises: taking the abbreviation data added with the suspicious label as a first crawler keyword, and taking the target full name data as a second crawler keyword; performing crawler search according to the first crawler keyword and the second crawler keyword to obtain a crawler search result, and taking the crawler search result as the result of the semantic analysis on the abbreviation data added with the suspicious label and the target full name data. An analysis unit is configured to perform data analysis on the target full name data to generate new abbreviation data if there is abbreviation data of the suspicious type in the M pieces of abbreviation data. A replacement unit is configured to replace the abbreviation data of the suspicious type in the M pieces of abbreviation data with the new abbreviation data in the target data set.

11. A computer device, comprising: A computer device comprises a processor, an input device, an output device and a memory, wherein the processor, the input device, the output device and the memory are connected to each other, the memory is configured to store a computer program, the computer program comprises program instructions, the processor is configured to invoke the program instructions, and execute the method according to any one of claims 1 to 9.

12. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a computer program, and the computer program comprises program instructions; when the program instructions are executed by a processor, the processor executes the method according to any one of claims 1 to 9.

13. A computer program product, comprising computer instructions stored in a computer readable storage medium; a processor of a computer device reads the computer instructions from the computer readable storage medium and executes the computer instructions, so that the computer device executes the method according to any one of claims 1 to 9.

Images