Data processing method and device, equipment and storage medium
By linking semantic phrases and tags of tagged data in the database, the problem of low efficiency in querying business results data was solved, and efficient and accurate data retrieval was achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- CHINA CONSTRUCTION BANK
- Filing Date
- 2023-04-17
- Publication Date
- 2026-06-09
AI Technical Summary
When business results data is large in volume and unstructured, query efficiency is low. Users find it difficult to match suitable data by directly entering search terms, resulting in high labor costs and high consumption of computing resources.
By parsing user query requests to obtain the target query time period and organization fields, historical query log data is read from the database, associated with predefined query semantic terms to determine the target query semantic terms, and associated with the tags of the tagged data to determine the target data.
It improves the accuracy and efficiency of queries, avoids useless queries caused by users frequently switching search terms, and reduces the consumption of human and computing resources.
Smart Images
Figure CN116361361B_ABST
Abstract
Description
Technical Field
[0001] This disclosure relates to the fields of big data technology and financial technology, and in particular to a data processing method, apparatus, device, medium and program product. Background Technology
[0002] In various data statistical analysis scenarios, staff will generate a huge amount of business results data based on multiple different business segments during statistical work. In the next statistical work, historical business results data will often be consulted for reference. At the same time, when summarizing projects, it is necessary to classify and statistically analyze the results in order to better analyze business results data from a global perspective.
[0003] In the process of developing this disclosure, it was discovered that due to the massive amount of business outcome data and the unstructured nature of this data, business personnel often need to expend considerable effort to search for the information they need from the vast amount of business outcome data accumulated over many years. This results in high labor costs, low work efficiency, and increased consumption of computer computing resources during the processing of massive amounts of data. Furthermore, users cannot predict suitable search terms in advance, making it difficult to match appropriate data based on the search terms directly entered by the user. Summary of the Invention
[0004] In view of the above problems, this disclosure provides a data processing method, apparatus, device, medium and program product.
[0005] One aspect of this disclosure provides a data processing method, comprising:
[0006] In response to a user query request, reference fields are obtained by parsing the user query request. The reference fields include at least the target query time period field and the target organization field to which the user belongs.
[0007] Based on the above target query period field and the above target organization field, retrieve the target historical query log data that was generated before the above target query period and is associated with the above target organization from the database;
[0008] The target historical query log data is associated with multiple predefined query semantic phrases to obtain the target query semantic phrases;
[0009] The target query semantic phrases mentioned above are associated with the data tags of multiple tagged data sets, so as to identify multiple target data sets from the multiple tagged data sets.
[0010] According to embodiments of this disclosure, the aforementioned target historical query log data is associated with multiple predefined query semantic phrases to obtain target query semantic phrases including:
[0011] The target historical query log data is matched with multiple of the above query semantic phrases to calculate the attention value corresponding to each of the above query semantic phrases;
[0012] The query semantic phrases whose attention value is greater than the first preset threshold are identified as the target query semantic phrases.
[0013] According to embodiments of this disclosure, the above-mentioned query semantic phrases include multi-level semantic units, and there are business association relationships among the above-mentioned multi-level semantic units;
[0014] By matching the aforementioned target historical query log data with multiple of the aforementioned query semantic phrases, the attention values corresponding to each of the aforementioned query semantic phrases are calculated as follows:
[0015] Determine the frequency of occurrence of each semantic unit at each level in the aforementioned query semantic phrase within the aforementioned target historical query log data;
[0016] Based on the frequency of occurrence of the semantic units at each level, the attention value corresponding to each of the above query semantic phrases is calculated.
[0017] According to embodiments of this disclosure, the above-mentioned query semantic phrases include a first type of query semantic phrases and a second type of query semantic phrases;
[0018] The aforementioned first type of query semantic phrases include multi-level first type semantic units, which include: control elements, control content, and control effects;
[0019] The aforementioned second type of query semantic phrases include multi-level second type semantic units, which include: risk subject, risk action, and risk object.
[0020] According to embodiments of this disclosure, the data tags of the tagged data include multi-level tag fields, and the query semantic phrases include multi-level semantic units;
[0021] The target query semantic terms mentioned above are associated with the data tags of multiple tagged data sets, so that multiple target data sets can be identified from the multiple tagged data sets, including:
[0022] The multi-level label field in the data label of the above-mentioned labeled data is semantically matched with the multi-level semantic unit in the above-mentioned target query semantic word group to calculate the matching degree value of each of the above-mentioned labeled data.
[0023] The tagged data whose matching degree value is greater than the second preset threshold are identified as the target data.
[0024] According to embodiments of this disclosure, the multi-level tag fields in the data tags of the tagged data are semantically matched with the multi-level semantic units in the target query semantic phrases, and the matching degree values of each of the tagged data are calculated as follows:
[0025] Each level of the multi-level label field is transformed to obtain multiple label feature vectors.
[0026] Each semantic unit in the above multi-level semantic unit is subjected to feature transformation to obtain multiple semantic feature vectors;
[0027] Calculate the similarity values between the above multiple label feature vectors and the above multiple semantic feature vectors;
[0028] Based on the similarity values between the aforementioned multiple label feature vectors and multiple semantic feature vectors, the matching degree value of each of the aforementioned labeled data is calculated.
[0029] According to embodiments of this disclosure, the above data processing method further includes:
[0030] The scoring values are obtained from the user's client and are used to characterize the user's satisfaction with the search results reflected in the above multiple sets of target data.
[0031] If the score is less than the third preset threshold, then the system will accept custom query semantic phrases from the client.
[0032] The aforementioned custom query semantic phrases are associated with the data labels of multiple tagged data sets, so that multiple custom query data sets can be identified from the aforementioned multiple tagged data sets.
[0033] According to embodiments of this disclosure, the above data processing method further includes:
[0034] Perform statistical analysis on the above multiple target data sets and output the statistical results;
[0035] The above statistical results are then visualized.
[0036] Another aspect of this disclosure provides a data processing apparatus, comprising: a first acquisition module, a generation module, a first association module, and a second association module. The first acquisition module is configured to, in response to a user query request, acquire reference fields by parsing the user query request. The reference fields include at least a target query time period field and a target organization field to which the user belongs. The generation module is configured to, based on the target query time period field and the target organization field, read target historical query log data from a database whose generation time is before the target query time period and which is associated with the target organization. The first association module is configured to associate the target historical query log data with multiple predefined query semantic phrases to obtain target query semantic phrases. The second association module is configured to associate the target query semantic phrases with data tags of multiple tagged data sets, thereby identifying multiple target data sets from the multiple tagged data sets.
[0037] Another aspect of this disclosure provides an electronic device comprising: one or more processors; and a memory for storing one or more programs, wherein, when the one or more programs are executed by the one or more processors, the one or more processors perform the data processing method described above.
[0038] Another aspect of this disclosure provides a computer-readable storage medium having executable instructions stored thereon, which, when executed by a processor, cause the processor to perform the data processing method described above.
[0039] Another aspect of this disclosure provides a computer program product, including a computer program that, when executed by a processor, implements the above-described data processing method.
[0040] According to the data processing method, apparatus, device, medium, and program product provided in this disclosure, in response to a user query request, at least the target query time period field and the target organization field to which the user belongs can be obtained by parsing the user query request. Based on the target query time period field and the target organization field, target historical query log data generated before the target query time period and associated with the target organization can be read from the database. The target historical query log data is associated with multiple predefined query semantic terms. The target query semantic terms can be obtained from the multiple predefined query semantic terms. Finally, the target query semantic terms are used as search terms and associated with the data tags of multiple tagged data. This allows multiple sets of target data that match the target query semantic terms to be identified from the multiple sets of tagged data as the information needed by the user. Since the target data is obtained by associating the target query semantic terms with the tagged data, multiple useless queries caused by the user frequently switching search terms can be avoided, further improving the accuracy and efficiency of the query. Attached Figure Description
[0041] The foregoing contents, as well as other objects, features, and advantages of this disclosure, will become clearer from the following description of embodiments with reference to the accompanying drawings, in which:
[0042] Figure 1 The illustrations depict application scenarios of data processing methods, apparatuses, devices, media, and program products according to embodiments of the present disclosure.
[0043] Figure 2 A flowchart illustrating a data processing method according to an embodiment of the present disclosure is shown schematically.
[0044] Figure 3 A schematic diagram of a record table for a first type of query semantic phrase according to an embodiment of the present disclosure is shown.
[0045] Figure 4 A schematic diagram of a record table for a second type of query semantic phrase according to an embodiment of the present disclosure is shown.
[0046] Figure 5 A flowchart illustrating the process of obtaining the matching degree value of tagged data according to an embodiment of the present disclosure is shown schematically.
[0047] Figure 6 A schematic block diagram of a data processing apparatus according to embodiments of the present disclosure is shown; and
[0048] Figure 7 A block diagram schematically illustrates an electronic device suitable for implementing a data processing method according to an embodiment of the present disclosure. Detailed Implementation
[0049] The embodiments of the present disclosure will now be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the disclosure. In the following detailed description, numerous specific details are set forth to provide a thorough understanding of the embodiments of the present disclosure for ease of explanation. However, it will be apparent that one or more embodiments may be practiced without these specific details. Furthermore, descriptions of well-known structures and techniques are omitted in the following description to avoid unnecessarily obscuring the concepts of the present disclosure.
[0050] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit this disclosure. The terms “comprising,” “including,” etc., as used herein indicate the presence of the stated features, steps, operations, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, or components.
[0051] All terms used herein (including technical and scientific terms) have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used herein are to be interpreted in a manner consistent with the context of this specification, and not in an idealized or overly rigid way.
[0052] When using expressions such as "at least one of A, B, and C", they should generally be interpreted in accordance with the meaning that is commonly understood by a person skilled in the art (e.g., "a system having at least one of A, B, and C" should include, but is not limited to, a system having A alone, a system having B alone, a system having C alone, a system having A and B, a system having A and C, a system having B and C, and / or a system having A, B, and C, etc.).
[0053] In the technical solutions disclosed herein, the collection, storage, use, processing, transmission, provision, disclosure, and application of data (including but not limited to user personal information) comply with the provisions of relevant laws and regulations, necessary confidentiality measures have been taken, and they do not violate public order and good morals.
[0054] In implementing this disclosure, it was discovered that related technologies only describe and record business results and store them in a simple structured manner, without establishing a professional business result data system. This results in disorganized and chaotic data classification. When dealing with massive amounts of data, only simple queries are possible, and it is impossible to clearly and efficiently classify and summarize information, or to form multi-perspective data analysis. Furthermore, users cannot predict suitable search terms in advance, making it difficult to match appropriate data based on the search terms directly entered by the user.
[0055] To this end, embodiments of this disclosure provide a data processing method, comprising: responding to a user query request, obtaining reference fields by parsing the user query request, the reference fields including at least a target query time period field and a target organization field to which the user belongs; based on the target query time period field and the target organization field, reading target historical query log data generated before the target query time period and associated with the target organization from the database; associating the target historical query log data with multiple predefined query semantic phrases to obtain target query semantic phrases; and associating the target query semantic phrases with data tags of multiple tagged data sets to identify multiple target data sets from the multiple tagged data sets.
[0056] Figure 1 The diagram illustrates an application scenario of data processing according to an embodiment of the present disclosure.
[0057] like Figure 1As shown, application scenario 100 according to this embodiment may include a first terminal device 101, a second terminal device 102, a third terminal device 103, a network 104, and a server 105. The network 104 serves as a medium for providing a communication link between the first terminal device 101, the second terminal device 102, the third terminal device 103, and the server 105. The network 104 may include various connection types, such as wired or wireless communication links, or fiber optic cables, etc.
[0058] Users can interact with server 105 via network 104 using at least one of the first terminal device 101, second terminal device 102, and third terminal device 103 to receive or send messages, etc. Various communication client applications can be installed on the first terminal device 101, second terminal device 102, and third terminal device 103, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social media platform software, etc. (for example only).
[0059] The first terminal device 101, the second terminal device 102, and the third terminal device 103 can be various electronic devices with displays and support web browsing, including but not limited to smartphones, tablets, laptops, and desktop computers.
[0060] For example, users can input query time period information and user department information through the first terminal device 101, the second terminal device 102, and the third terminal device 103 to generate a user query request.
[0061] Server 105 can be a server that provides various services, such as a backend management server that supports websites browsed by users using the first terminal device 101, the second terminal device 102, and the third terminal device 103 (this is just an example). The backend management server can analyze and process data such as received user requests, and feed back the processing results (such as web pages, information, or data obtained or generated according to user requests) to the terminal devices.
[0062] For example, server 105 can respond to user query requests, parse the user query requests to obtain reference fields, which include at least the target query time period field and the target organization field to which the user belongs; and based on the target query time period field and the target organization field, read the target historical query log data generated before the target query time period and associated with the target organization from the database; then associate the target historical query log data with multiple predefined query semantic terms to obtain the target query semantic terms; finally, associate the target query semantic terms with the data tags of multiple tagged data to identify multiple target data from the multiple tagged data.
[0063] It should be noted that the data processing method provided in this embodiment can generally be executed by server 105. Correspondingly, the data processing device provided in this embodiment can generally be located in server 105. The data processing method provided in this embodiment can also be executed by a server or server cluster that is different from server 105 and capable of communicating with the first terminal device 101, the second terminal device 102, the third terminal device 103, and / or server 105. Correspondingly, the data processing device provided in this embodiment can also be located in a server or server cluster that is different from server 105 and capable of communicating with the first terminal device 101, the second terminal device 102, the third terminal device 103, and / or server 105.
[0064] It should be understood that Figure 1 The number of terminal devices, networks, and servers shown is merely illustrative. Depending on implementation needs, any number of terminal devices, networks, and servers can be included.
[0065] The following will be based on Figure 1 The described scene, through Figures 2-5 The data processing method of the disclosed embodiments will be described in detail.
[0066] Figure 2 A flowchart illustrating a data processing method according to an embodiment of the present disclosure is shown schematically.
[0067] like Figure 2 As shown, the method 200 includes operations S210 to S240.
[0068] In operation S210, in response to a user query request, reference fields are obtained by parsing the user query request. The reference fields include at least the target query time period field and the target organization field to which the user belongs.
[0069] According to embodiments of this disclosure, when a user needs to find the information they require from a large number of business results, and the user cannot predict suitable search terms in advance, the user can input the desired time period and their department / organization information to generate a user query request. By parsing the user query request, reference fields are obtained. These reference fields represent fields corresponding to the user's input information obtained from parsing the query request. Related historical query data can be retrieved from the database based on these reference fields.
[0070] According to embodiments of this disclosure, the reference field may include at least a target query time period field and a target organization field to which the user belongs. The reference field may also include a responsibility field for the target organization to which the user belongs. The target query time period field may represent the query time period information entered by the user, and the target organization field may represent the department or organization to which the user belongs.
[0071] In operation S220, based on the target query time period field and the target organization field, target historical query log data that was generated before the target query time period and is associated with the target organization is read from the database.
[0072] According to embodiments of this disclosure, historical query data generated before the target query period is read from the database based on the target query period field. Then, based on this historical query data, data associated with the target organization is further read from this historical query data according to the target organization field, thereby obtaining the target historical query log data. For example, if a user inputs a query period from June 2020 to June 2021, and the user's department / organization information is a first-level branch, then the historical query data of the first-level branch within the period from June 2020 to June 2021 can be read from the database.
[0073] According to embodiments of this disclosure, historical query data related to the information needed by the user can be read from the database based on reference fields obtained by parsing the user's query request. The historical query data in the database may include historical query terms and time information matching those terms. For example, a query term might be "first-level branch + waste + cash," and the time information matching that query term might be July 1, 2020, 13:00:00.
[0074] In operation S230, the target historical query log data is associated with multiple predefined query semantic terms to obtain the target query semantic terms.
[0075] According to embodiments of this disclosure, predefined query semantic phrases can represent query semantic phrases obtained by dividing query terms according to certain rules, and are configured in advance according to certain rules. Historical query terms in the database and predefined query semantic phrases satisfy the same division rules.
[0076] According to embodiments of this disclosure, target historical query log data read from a database is associated with multiple predefined query semantic phrases, and a target query semantic phrase can be obtained from these predefined phrases. The target query semantic phrase can represent query terms that match the information needed by the user.
[0077] In operation S240, the target query semantic phrase is associated with the data labels of multiple tagged data sets so as to identify multiple target data sets from the multiple tagged data sets.
[0078] According to embodiments of this disclosure, tagged data can characterize data labeled with query semantic phrases, and data labels can characterize query semantic phrases.
[0079] According to embodiments of this disclosure, by associating target query semantic phrases as query terms with data tags of multiple tagged data sets, multiple sets of target data matching the target query semantic phrases can be determined from the multiple tagged data sets. The target data can represent the information needed by the user based on the target query semantic phrases.
[0080] According to embodiments of this disclosure, in response to a user query request, at least the target query time period field and the target organization field to which the user belongs can be obtained by parsing the user query request. Based on the target query time period field and the target organization field, target historical query log data generated before the target query time period and associated with the target organization can be read from the database. The target historical query log data is then associated with multiple predefined query semantic phrases. The target query semantic phrase can be obtained from the multiple predefined query semantic phrases. Finally, the target query semantic phrase is used as a search term and associated with the data tags of multiple tagged data sets. This allows multiple sets of target data matching the target query semantic phrase to be determined from the multiple tagged data sets as the information needed by the user. Since the target data is obtained by associating the target query semantic phrase with the tagged data sets, multiple useless queries caused by the user frequently switching search terms can be avoided, further improving the accuracy and efficiency of the query.
[0081] According to embodiments of this disclosure, associating target historical query log data with multiple predefined query semantic phrases to obtain target query semantic phrases includes: matching target historical query log data with multiple query semantic phrases to calculate the attention value corresponding to each query semantic phrase; and determining query semantic phrases with attention values greater than a first preset threshold as target query semantic phrases.
[0082] According to embodiments of this disclosure, multiple predefined query semantic phrases can be stored in a table. By matching the target historical query log data with the multiple query semantic phrases obtained from the table, the attention value corresponding to each query semantic phrase can be obtained.
[0083] According to embodiments of this disclosure, the attention values corresponding to each obtained query semantic word group can be compared with a first preset threshold, and the query semantic word groups with attention values greater than the first preset threshold are determined as target query semantic word groups. For example, the attention values corresponding to each query semantic word group can be 300, 200, and 280, respectively. If the first preset threshold is set to 220, then the query semantic word groups with attention values of 300 and 280 are determined as target query semantic word groups.
[0084] According to embodiments of this disclosure, the attention value of each query semantic term is calculated, and the attention value is compared with a first preset threshold. Query semantic terms with attention values greater than the first preset threshold are identified as target query semantic terms. This facilitates subsequent queries using the target query semantic terms as the search terms for the information needed by the user, avoiding multiple useless queries caused by the user frequently switching search terms, and improving the accuracy and efficiency of the query.
[0085] According to embodiments of this disclosure, matching target historical query log data with multiple query semantic phrases and calculating the attention value corresponding to each query semantic phrase includes: determining the number of times each level of semantic unit in the query semantic phrase appears in the target historical query log data; and calculating the attention value corresponding to each query semantic phrase based on the number of times each level of semantic unit appears.
[0086] According to embodiments of this disclosure, query semantic phrases may include multi-level semantic units, and there are business relationships between the multi-level semantic units.
[0087] According to embodiments of this disclosure, the frequency of occurrence of each semantic unit at each level within a query semantic term in the obtained target historical query log data is determined. Based on the frequency of occurrence of each semantic unit at each level within the target historical query log data, the attention value corresponding to each query semantic term can be calculated. For example, for a query semantic term "first-level branch + waste + cash", the semantic units at each level within this query semantic term can be represented as "first-level branch", "waste", and "cash", respectively. By searching for the frequency of occurrence of "first-level branch", "waste", and "cash" in the target historical query log data, it can be found that "first-level branch" appears 100 times, "waste" appears 50 times, and "cash" appears 50 times. Therefore, the frequency of occurrence of "first-level branch", "waste", and "cash" can be added together as the attention value corresponding to this query semantic term, i.e., the attention value is 200. Similarly, the sum of the occurrences of each semantic unit in a query semantic phrase can be a preset multiple of the corresponding attention value. For example, if the preset multiple is 50, and the sum of the occurrences of each semantic unit in a query semantic phrase is 300, then the corresponding attention value can be 6.
[0088] According to embodiments of this disclosure, when the attention value can represent the sum of the occurrences of semantic units at each level in the query semantic phrase, the first preset threshold can be set to a value corresponding to the attention value. For example, if the attention values of the query semantic phrase are 300 and 200 respectively, then the first preset threshold can be set to 260. When the sum of the occurrences of semantic units at each level in the query semantic phrase can be a preset multiple of the corresponding attention value, the first preset threshold can be set to a value corresponding to the attention value. For example, the preset multiple can be set to 50. If the sum of the occurrences of semantic units at each level in the query semantic phrase is 300, then the corresponding attention value can be 6, and the first preset threshold can be set to 5.
[0089] According to embodiments of this disclosure, by determining the number of times each level of semantic unit in the query semantic term group appears in the target historical query log data, the attention value corresponding to each query semantic term group is calculated, so that the target query semantic term group can be determined from multiple query semantic term groups as subsequent query retrieval terms based on the calculated attention value.
[0090] According to embodiments of this disclosure, query semantic phrases may include a first type of query semantic phrase and a second type of query semantic phrase. The first type of query semantic phrase may include multi-level first-class semantic units, which may include: control elements, control content, and control effects. The second type of query semantic phrase may include multi-level second-class semantic units, which may include: risk subject, risk action, and risk object.
[0091] According to embodiments of this disclosure, the first type of query semantic phrases are obtained by dividing according to the rules of control elements, control content, and control effects, and the second type of query semantic phrases are obtained by dividing according to the rules of risk subject, risk action, and risk object.
[0092] Figure 3 A schematic diagram of a configuration table for a first type of query semantic phrase according to an embodiment of the present disclosure is shown.
[0093] like Figure 3 The table shown is a configuration table 300 for the first type of query semantic phrases. The first type of query semantic phrases can represent semantic phrases related to internal control deficiencies. The first type of query semantic phrases can include multi-level first-type semantic units, which can include three parts: control elements, control content, and control effects. In other words, the first type of query semantic phrases can be composed of these three parts.
[0094] According to embodiments of this disclosure, control elements may include internal environment, internal management, normative controls, property rights controls, and technological controls. Control content may include organizational structure, departmental division of labor, collaborative processes, policy implementation, and strategy execution. Control effects may include incoordination, superficiality, irrationality, non-standardization, and inadequate implementation.
[0095] According to embodiments of this disclosure, such as Figure 3 As shown, the first type of query semantic phrase can be represented by a phrase in the control element part, a phrase in the control content part, and a phrase in the control effect part. For example, the first type of query semantic phrase can include "internal management + departmental division of labor + lack of coordination", or it can include "internal management + policy implementation + inadequate implementation".
[0096] Figure 4 A schematic diagram of a configuration table for a second type of query semantic phrase according to an embodiment of the present disclosure is shown.
[0097] like Figure 4 The table shown is a configuration table 400 for the second type of query semantic phrases. The second type of query semantic phrases can represent semantic phrases related to important risk events. The second type of query semantic phrases can include multi-level second-type semantic units, which can include three parts: risk subject, risk action, and risk object. In other words, the second type of query semantic phrases can be composed of these three parts.
[0098] According to embodiments of this disclosure, risk subjects may include head office departments, first-tier branches, second-tier branches, and grassroots institutions. Risk actions may include waste, misuse, misappropriation, and misreporting. Risk objects may include cash, projects, products, and systems.
[0099] According to embodiments of this disclosure, such as Figure 4 As shown, the second type of query semantic phrase can be represented by a phrase in the risk subject part, a phrase in the risk action part, and a phrase in the risk object part. For example, the second type of query semantic phrase can include "first-level branch + waste + cash" or "second-level branch + waste + cash".
[0100] According to embodiments of this disclosure, the first type of query semantic terms are divided according to the rules of control elements, control content, and control effects, and the second type of query semantic terms are divided according to the rules of risk subject, risk action, and risk object. This can make the query search terms structured, making it easier for users to find the information they need.
[0101] According to embodiments of this disclosure, associating a target query semantic phrase with data tags of multiple tagged data sets to determine multiple target data sets from the multiple tagged data sets includes: semantically matching the multi-level tag fields in the data tags of the tagged data sets with the multi-level semantic units in the target query semantic phrase phrases to calculate the matching degree value of each tagged data set; and determining the tagged data sets with matching degree values greater than a second preset threshold as target data sets.
[0102] According to embodiments of this disclosure, the data tags of the tagged data may include multi-level tag fields, and the query semantic phrases may include multi-level semantic units.
[0103] According to embodiments of this disclosure, tagged data can represent data labeled with query semantic phrases, that is, by querying with query semantic phrases, the corresponding tagged data labeled with the query semantic phrases can be obtained.
[0104] According to embodiments of this disclosure, multi-level tag fields can represent multi-level semantic units. Each level of the multi-level semantic unit in the target query semantic phrase is matched with each level of the multi-level tag field in the data tags of the tagged data, and the matching degree value between each tagged data and the target query semantic phrase is calculated. For example, the target query semantic phrase can be represented as "first-level branch + waste + cash". The semantic unit "first-level branch" in the target query semantic phrase is semantically matched with the corresponding risk subject part in the multi-level tag field of the data tags of each tagged data; the semantic unit "waste" in the target query semantic phrase is semantically matched with the corresponding risk action part in the multi-level tag field of the data tags of each tagged data; and the semantic unit "cash" in the target query semantic phrase is semantically matched with the corresponding risk object part in the multi-level tag field of the data tags of each tagged data. Based on the results of the above three matchings, the matching degree value of each tagged data can be obtained.
[0105] According to embodiments of this disclosure, the matching degree value corresponding to the tagged data is compared with a second preset threshold, and the tagged data with a matching degree value greater than the second preset threshold is determined as target data. The second preset threshold is manually set and is used to determine the target data that can be obtained by querying using the target query semantic phrase as the search term, wherein the target data can represent the information required by the user.
[0106] According to the embodiments of this disclosure, the multi-level tag field in the data tags of the tagged data is semantically matched with the multi-level semantic unit in the target query semantic phrase. The matching degree value of each tagged data can be calculated, and the matching degree between the data tags in multiple tagged data and the target query semantic phrase can be determined. The tagged data with a matching degree value greater than the second preset threshold is determined as the target data by using the target query semantic phrase as the query retrieval term. This realizes the information needed by the user, i.e., the target data, obtained by querying with the target query semantic phrase as the query retrieval term.
[0107] Figure 5 A flowchart illustrating the process of obtaining the matching degree value of tagged data according to an embodiment of the present disclosure is shown.
[0108] like Figure 5 As shown, the method 500 includes operations S510 to S540.
[0109] In operation S510, feature transformation is performed on each level of the multi-level label field to obtain multiple label feature vectors.
[0110] According to embodiments of this disclosure, a multi-level label field can represent multi-level semantic units, and thus the multi-level label field may include multi-level first-class semantic units and multi-level second-class semantic units. By performing feature transformation on each level of the label field in the multi-level label field, multiple label feature vectors corresponding to each level of the label field can be obtained. For example, when the multi-level label field can represent "first-level branch + waste + cash," performing feature transformation on each level of the label field in the multi-level label field can obtain a semantic feature vector corresponding to "first-level branch," a semantic feature vector corresponding to "waste," and a semantic feature vector corresponding to "cash."
[0111] In the S520 operation, the semantic units at each level in the multi-level semantic unit are transformed to obtain multiple semantic feature vectors.
[0112] According to embodiments of this disclosure, a multi-level semantic unit can be a multi-level first-class semantic unit and a multi-level second-class semantic unit. By performing feature transformation on each level of the multi-level semantic unit, a feature vector corresponding to each level can be obtained. For example, when the multi-level semantic unit can represent "first-level branch + misuse + cash", performing feature transformation on each level of the multi-level semantic unit can obtain a semantic feature vector corresponding to "first-level branch", a semantic feature vector corresponding to "misuse", and a semantic feature vector corresponding to "cash".
[0113] In operation S530, the similarity values between multiple label feature vectors and multiple semantic feature vectors are calculated.
[0114] According to embodiments of this disclosure, similarity values are calculated for corresponding feature vectors between multiple label feature vectors and multiple semantic feature vectors. Both the multi-level label field and the multi-level semantic unit can be divided into three parts: calculating the similarity value between the label feature vector corresponding to the first part of the multi-level label field and the semantic feature vector corresponding to the first part of the multi-level semantic unit; calculating the similarity value between the label feature vector corresponding to the second part of the multi-level label field and the semantic feature vector corresponding to the second part of the multi-level semantic unit; and calculating the similarity value between the label feature vector corresponding to the third part of the multi-level label field and the semantic feature vector corresponding to the third part of the multi-level semantic unit. For example, when a multi-level label field can represent "first-level branch + waste + cash" and a multi-level semantic unit can represent "first-level branch + misuse + cash", calculate the similarity value between the label feature vector corresponding to the first part "first-level branch" in the multi-level label field and the semantic feature vector corresponding to the first part "first-level branch" in the multi-level semantic unit; calculate the similarity value between the label feature vector corresponding to the second part "waste" in the multi-level label field and the semantic feature vector corresponding to the second part "misuse" in the multi-level semantic unit; calculate the similarity value between the label feature vector corresponding to the third part "cash" in the multi-level label field and the semantic feature vector corresponding to the third part "cash" in the multi-level semantic unit.
[0115] In operation S540, the matching degree value of each labeled data is calculated based on the similarity values between multiple label feature vectors and multiple semantic feature vectors.
[0116] According to embodiments of this disclosure, the matching degree value corresponding to each tagged data can be obtained based on the similarity values corresponding to the three parts of each multi-level label field and multi-level semantic unit.
[0117] According to embodiments of this disclosure, similarity values between multiple label feature vectors and multiple semantic feature vectors can be calculated using the label feature vectors corresponding to each level of label fields in a multi-level label field and the semantic feature vectors corresponding to each level of semantic units in a multi-level semantic unit. Based on the similarity values between the multiple label feature vectors and multiple semantic feature vectors, the matching degree value of each tagged data can be calculated so that the target data can be determined from the multiple tagged data based on the matching degree value of the tagged data.
[0118] According to embodiments of this disclosure, the above data processing method further includes: obtaining a scoring value from a user client, the scoring value being used to characterize the user's satisfaction with the search results reflected in multiple sets of target data; and, if the scoring value is less than a third preset threshold, receiving a custom query semantic phrase from the client, and associating the custom query semantic phrase with the data tags of multiple sets of labeled data, so as to determine multiple sets of custom query data from the multiple sets of labeled data.
[0119] According to embodiments of this disclosure, users can obtain multiple search results based on target data by using target query semantic phrases as search terms, and can obtain a score value based on the user's satisfaction with the search results.
[0120] According to embodiments of this disclosure, the third preset threshold can represent the user's minimum level of satisfaction with the search results. Custom query semantic phrases can represent query semantic phrases defined by the user as needed. When the score is less than the third preset threshold, i.e., the user is dissatisfied with the search results reflected by multiple sets of target data, a custom query semantic phrase from the client can be received. That is, the user can define a custom query semantic phrase and associate it with the data tags of multiple tagged data sets, so as to determine multiple sets of custom query data from the multiple tagged data sets. The custom query data can represent the search results obtained by using the custom query semantic phrase as the search term.
[0121] According to embodiments of this disclosure, associating a custom query semantic phrase with data tags of multiple tagged data sets, so that determining multiple custom query data sets from multiple tagged data sets may include: semantically matching the multi-level tag fields in the data tags of the tagged data sets with the multi-level semantic units in the custom query semantic phrase phrases, and calculating the matching degree value of each tagged data set; and determining the tagged data sets with matching degree values greater than a second preset threshold as custom query data sets.
[0122] According to embodiments of this disclosure, using target query semantic phrases as search terms can yield multiple search results reflecting target data. If a user is not satisfied with the search results, the user can use custom query semantic phrases as search terms to obtain custom query data.
[0123] According to embodiments of this disclosure, the above data processing method further includes: performing statistical analysis on multiple sets of target data and outputting statistical results; and visualizing the statistical results.
[0124] According to embodiments of this disclosure, statistics are performed on multiple sets of target data obtained based on target query semantic phrases, and the statistical results can be visualized through a visualization interface. Similarly, statistical results of custom query results obtained by users based on custom query semantic phrases can also be visualized.
[0125] According to embodiments of this disclosure, visualizing the statistical results makes it easier for users to intuitively and clearly understand the statistical results of the query, and to judge whether they are satisfied based on the statistical results.
[0126] Based on the above data processing method, this disclosure also provides a data processing apparatus. The following will be combined with... Figure 6 The device is described in detail.
[0127] Figure 6 A schematic block diagram of a data processing apparatus according to an embodiment of the present disclosure is shown.
[0128] like Figure 6 As shown, the data processing device 600 in this embodiment includes a first acquisition module 610, a generation module 620, a first association module 630, and a second association module 640.
[0129] The first acquisition module 610 is used to respond to a user query request by parsing the user query request to obtain reference fields. The reference fields include at least a target query time period field and a target organization field to which the user belongs. In one embodiment, the first acquisition module 610 can be used to perform the operation S210 described above, which will not be repeated here.
[0130] The generation module 620 is used to read target historical query log data from the database based on the target query time period field and the target organization field, where the generation time is before the target query time period and is associated with the target organization. In one embodiment, the generation module 620 can be used to perform the operation S220 described above, which will not be repeated here.
[0131] The first association module 630 is used to associate the target historical query log data with multiple predefined query semantic phrases to obtain the target query semantic phrases. In one embodiment, the first association module 630 can be used to perform the operation S230 described above, which will not be repeated here.
[0132] The second association module 640 is used to associate the target query semantic phrase with the data tags of multiple tagged data sets, so as to determine multiple target data sets from the multiple tagged data sets. In one embodiment, the second association module 640 can be used to perform the operation S240 described above, which will not be repeated here.
[0133] According to embodiments of this disclosure, the first association module 630 includes a first obtaining unit and a first determining unit.
[0134] The first acquisition unit is used to match the target historical query log data with multiple query semantic phrases and calculate the attention value corresponding to each query semantic phrase.
[0135] The first determining unit is used to determine the query semantic phrases with a focus value greater than a first preset threshold as target query semantic phrases.
[0136] According to embodiments of this disclosure, the first obtaining unit includes a determining subunit and an obtaining subunit.
[0137] Determine sub-units to determine the frequency of occurrence of each level of semantic units in the query semantic phrase in the target historical query log data.
[0138] Obtain sub-units, which are used to calculate the attention value corresponding to each query semantic phrase based on the occurrence frequency of each level of semantic unit.
[0139] According to embodiments of this disclosure, the second association module 640 includes a second obtaining unit and a second determining unit.
[0140] The second obtaining unit is used to perform semantic matching between the multi-level label field in the data label of the labeled data and the multi-level semantic unit in the target query semantic phrase, and calculate the matching degree value of each labeled data.
[0141] The second determining unit is used to determine the tagged data whose matching degree value is greater than the second preset threshold as target data.
[0142] According to embodiments of this disclosure, the second obtaining unit includes a first conversion subunit, a second conversion subunit, a first calculation subunit, and a second calculation subunit.
[0143] The first transformation subunit is used to perform feature transformation on each level of the label field in the multi-level label field to obtain multiple label feature vectors.
[0144] The second transformation subunit is used to transform the features of each level of semantic units in the multi-level semantic unit to obtain multiple semantic feature vectors.
[0145] The first computational subunit is used to calculate the similarity value between multiple label feature vectors and multiple semantic feature vectors.
[0146] The second calculation subunit is used to calculate the matching degree value of each labeled data based on the similarity values between multiple label feature vectors and multiple semantic feature vectors.
[0147] According to embodiments of this disclosure, the data processing apparatus further includes a second acquisition module, a receiving module, and a third association module.
[0148] The second acquisition module is used to obtain the scoring values from the user's client. The scoring values are used to characterize the user's satisfaction with the search results reflected in multiple sets of target data.
[0149] The receiving module is used to receive custom query semantic phrases from the client when the score value is less than a third preset threshold.
[0150] The third association module is used to associate custom query semantic phrases with data labels of multiple tagged data sets, so as to identify multiple custom query data sets from multiple tagged data sets.
[0151] According to embodiments of this disclosure, the data processing apparatus further includes an output module and a display module.
[0152] The output module is used to perform statistical analysis on multiple target data sets and output the statistical results.
[0153] The display module is used to visualize the statistical results.
[0154] According to embodiments of this disclosure, any plurality of modules among the first acquisition module 610, generation module 620, first association module 630, and second association module 640 can be combined into one module, or any one of these modules can be split into multiple modules. Alternatively, at least part of the functionality of one or more of these modules can be combined with at least part of the functionality of other modules and implemented in one module. According to embodiments of this disclosure, at least one of the first acquisition module 610, generation module 620, first association module 630, and second association module 640 can be at least partially implemented as hardware circuitry, such as a field-programmable gate array (FPGA), a programmable logic array (PLA), a system-on-a-chip, a system-on-a-substrate, a system-on-package, an application-specific integrated circuit (ASIC), or implemented in hardware or firmware by any other reasonable means of integrating or packaging the circuitry, or implemented in any one of software, hardware, and firmware methods, or in a suitable combination of any of these methods. Alternatively, at least one of the first acquisition module 610, generation module 620, first association module 630 and second association module 640 can be at least partially implemented as a computer program module, which can perform corresponding functions when the computer program module is run.
[0155] Figure 7 A block diagram schematically illustrates an electronic device suitable for implementing a data processing method according to an embodiment of the present disclosure.
[0156] like Figure 7As shown, an electronic device 700 according to an embodiment of the present disclosure includes a processor 701, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 702 or a program loaded from a storage portion 708 into a random access memory (RAM) 703. The processor 701 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or an associated chipset and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 701 may also include onboard memory for caching purposes. The processor 701 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of the present disclosure.
[0157] RAM 703 stores various programs and data required for the operation of electronic device 700. Processor 701, ROM 702, and RAM 703 are interconnected via bus 704. Processor 701 performs various operations of the method flow according to embodiments of the present disclosure by executing programs in ROM 702 and / or RAM 703. It should be noted that the programs may also be stored in one or more memories other than ROM 702 and RAM 703. Processor 701 may also perform various operations of the method flow according to embodiments of the present disclosure by executing programs stored in said one or more memories.
[0158] According to embodiments of this disclosure, the electronic device 700 may further include an input / output (I / O) interface 705, which is also connected to a bus 704. The electronic device 700 may also include one or more of the following components connected to the input / output (I / O) interface 705: an input section 706 including a keyboard, mouse, etc.; an output section 707 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 708 including a hard disk, etc.; and a communication section 709 including a network interface card such as a LAN card, modem, etc. The communication section 709 performs communication processing via a network such as the Internet. A drive 710 is also connected to the input / output (I / O) interface 705 as needed. A removable medium 711, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on the drive 710 as needed so that computer programs read from it can be installed into the storage section 708 as needed.
[0159] This disclosure also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments; or it may exist independently and not assembled into the device / apparatus / system. The computer-readable storage medium carries one or more programs that, when executed, implement the method according to the embodiments of this disclosure.
[0160] According to embodiments of this disclosure, the computer-readable storage medium may be a non-volatile computer-readable storage medium, such as, but not limited to: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this disclosure, the computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. For example, according to embodiments of this disclosure, the computer-readable storage medium may include ROM 702 and / or RAM 703 and / or one or more memories other than ROM 702 and RAM 703 described above.
[0161] Embodiments of this disclosure also include a computer program product comprising a computer program containing program code for performing the methods shown in the flowchart. When the computer program product is run on a computer system, the program code is used to enable the computer system to implement the data processing methods provided in the embodiments of this disclosure.
[0162] When the computer program is executed by the processor 701, it performs the functions defined in the system / apparatus of this disclosure embodiments. According to embodiments of this disclosure, the systems, apparatuses, modules, units, etc., described above can be implemented by computer program modules.
[0163] In one embodiment, the computer program may rely on a tangible storage medium such as an optical storage device or a magnetic storage device. In another embodiment, the computer program may also be transmitted and distributed in the form of signals over a network medium, and may be downloaded and installed via the communication section 709, and / or installed from a removable medium 711. The program code contained in the computer program can be transmitted using any suitable network medium, including but not limited to: wireless, wired, etc., or any suitable combination thereof.
[0164] In such an embodiment, the computer program can be downloaded and installed from a network via the communication section 709, and / or installed from the removable medium 711. When the computer program is executed by the processor 701, it performs the functions defined in the system of this disclosure embodiment. According to embodiments of this disclosure, the systems, devices, apparatuses, modules, units, etc., described above can be implemented by computer program modules.
[0165] According to embodiments of this disclosure, program code for executing the computer programs provided in embodiments of this disclosure can be written in any combination of one or more programming languages. Specifically, these computational programs can be implemented using high-level procedural and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages include, but are not limited to, languages such as Java, C++, Python, "C", or similar programming languages. The program code can execute entirely on the user's computing device, partially on the user's device, partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).
[0166] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.
[0167] Those skilled in the art will understand that the features described in the various embodiments and / or claims of this disclosure can be combined or combined in various ways, even if such combinations or combinations are not explicitly described in this disclosure. In particular, the features described in the various embodiments and / or claims of this disclosure can be combined or combined in various ways without departing from the spirit and teachings of this disclosure. All such combinations and / or combinations fall within the scope of this disclosure.
[0168] The embodiments of this disclosure have been described above. However, these embodiments are for illustrative purposes only and are not intended to limit the scope of this disclosure. Although various embodiments have been described above, this does not mean that the measures in the various embodiments cannot be used advantageously in combination. The scope of this disclosure is defined by the appended claims and their equivalents. Various substitutions and modifications can be made by those skilled in the art without departing from the scope of this disclosure, and all such substitutions and modifications should fall within the scope of this disclosure.
Claims
1. A data processing method, comprising: In response to a user query request, reference fields are obtained by parsing the user query request. The reference fields include at least the target query time period field and the target organization field to which the user belongs. Based on the target query period field and the target organization field, retrieve target historical query log data from the database that was generated before the target query period and is associated with the target organization; Associating the target historical query log data with multiple predefined query semantic phrases to obtain target query semantic phrases includes: matching the target historical query log data with multiple query semantic phrases to calculate the attention value corresponding to each query semantic phrase; identifying query semantic phrases with attention values greater than a first preset threshold as target query semantic phrases; wherein, the query semantic phrase includes multi-level semantic units, and there are business relationships between the multi-level semantic units; matching the target historical query log data with multiple query semantic phrases to calculate the attention value corresponding to each query semantic phrase includes: determining the number of times each level of semantic unit in the query semantic phrase appears in the target historical query log data; calculating the attention value corresponding to each query semantic phrase based on the number of times each level of semantic unit appears; The target query semantic phrase is associated with the data tags of multiple tagged data sets, so as to identify multiple target data sets from the multiple tagged data sets.
2. The method according to claim 1, wherein: The query semantic phrases include a first type of query semantic phrases and a second type of query semantic phrases; The first type of query semantic phrase includes multi-level first-class semantic units, and the multi-level first-class semantic units include: control elements, control content, and control effects; The second type of query semantic phrases includes multi-level second-type semantic units, which include: risk subject, risk action, and risk object.
3. The method according to claim 1, wherein, The data tags of the labeled data include multi-level tag fields, and the query semantic phrases include multi-level semantic units; Associating the target query semantic phrase with the data tags of multiple tagged data sets, so that determining multiple target data sets from the multiple tagged data sets includes: The multi-level tag field in the data tag of the labeled data is semantically matched with the multi-level semantic unit in the target query semantic word group to calculate the matching degree value of each labeled data. The tagged data whose matching degree value is greater than the second preset threshold is determined as the target data.
4. The method according to claim 3, wherein, The multi-level tag field in the data tags of the labeled data is semantically matched with the multi-level semantic unit in the target query semantic term group, and the matching degree value of each labeled data is calculated, including: Each level of the multi-level label field is subjected to feature transformation to obtain multiple label feature vectors; Each semantic unit in the multi-level semantic unit is subjected to feature transformation to obtain multiple semantic feature vectors; Calculate the similarity value between the plurality of label feature vectors and the plurality of semantic feature vectors; The matching degree value of each of the labeled data is calculated based on the similarity values between the multiple label feature vectors and the multiple semantic feature vectors.
5. The method according to any one of claims 1-4, further comprising: The scoring values are obtained from the user's client and are used to characterize the user's satisfaction with the search results reflected in the multiple sets of target data. If the score value is less than the third preset threshold, a custom query semantic phrase from the client is received; The custom query semantic phrases are associated with the data tags of multiple tagged data sets, so that multiple custom query data sets can be determined from the multiple tagged data sets.
6. The method according to any one of claims 1-4, further comprising: Perform statistical analysis on the multiple target data sets and output the statistical results; The statistical results are then visualized.
7. A data processing apparatus, comprising: The first acquisition module is used to respond to a user query request by parsing the user query request to obtain reference fields, the reference fields including at least a target query time period field and a target organization field to which the user belongs; The generation module is used to read target historical query log data associated with the target institution from the database based on the target query period field and the target institution field. The first association module is used to associate the target historical query log data with multiple predefined query semantic phrases to obtain the target query semantic phrases; The second association module is used to associate the target query semantic phrase with the data tags of multiple tagged data, so as to determine multiple target data from the multiple tagged data; The first association module includes a first obtaining unit and a first determining unit; the first obtaining unit is used to match the target historical query log data with multiple query semantic word groups and calculate the attention value corresponding to each query semantic word group; the first determining unit is used to determine the query semantic word groups with attention values greater than a first preset threshold as target query semantic word groups; The query semantic phrases include multi-level semantic units, and there are business relationships between the multi-level semantic units; The first obtaining unit includes a determining subunit and an obtaining subunit; the determining subunit is used to determine the number of times each level of semantic unit in the query semantic phrase appears in the target historical query log data; the obtaining subunit is used to calculate the attention value corresponding to each query semantic phrase based on the number of times each level of semantic unit appears.
8. An electronic device, comprising: One or more processors; Storage device for storing one or more programs. Wherein, when the one or more programs are executed by the one or more processors, the one or more processors perform the method according to any one of claims 1 to 6.
9. A computer-readable storage medium having stored thereon executable instructions that, when executed by a processor, cause the processor to perform the method according to any one of claims 1 to 6.
10. A computer program product comprising a computer program that, when executed by a processor, implements the method according to any one of claims 1 to 6.