CN117540107A - Data processing method and device, electronic equipment and storage medium - Google Patents
Data processing method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN117540107A CN117540107A CN202410030664.6A CN202410030664A CN117540107A CN 117540107 A CN117540107 A CN 117540107A CN 202410030664 A CN202410030664 A CN 202410030664A CN 117540107 A CN117540107 A CN 117540107A
- Authority
- CN
- China
- Prior art keywords
- data
- attribute
- intention
- attribute data
- input
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000003672 processing method Methods 0.000 title abstract description 18
- 238000000034 method Methods 0.000 claims abstract description 36
- 238000012545 processing Methods 0.000 claims abstract description 10
- 238000004458 analytical method Methods 0.000 claims description 36
- 230000000007 visual effect Effects 0.000 claims description 17
- 238000000605 extraction Methods 0.000 claims description 8
- 238000013434 data augmentation Methods 0.000 claims 1
- 238000012549 training Methods 0.000 description 34
- 238000012706 support-vector machine Methods 0.000 description 11
- 238000010586 diagram Methods 0.000 description 10
- 238000012800 visualization Methods 0.000 description 10
- 238000013075 data extraction Methods 0.000 description 8
- 238000006243 chemical reaction Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 230000003993 interaction Effects 0.000 description 2
- 238000003062 neural network model Methods 0.000 description 2
- 230000011218 segmentation Effects 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/903—Querying
- G06F16/9032—Query formulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/903—Querying
- G06F16/90335—Query processing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/904—Browsing; Visualisation therefor
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9538—Presentation of query results
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/10—Machine learning using kernel methods, e.g. support vector machines [SVM]
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Computing Systems (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The application provides a data processing method, a data processing device, electronic equipment and a storage medium, and relates to the technical field of computers; the method comprises the following steps: receiving input data input by a target object; extracting attribute data from the input data to obtain at least one attribute data; determining an intention category corresponding to the attribute data, wherein the intention category comprises at least one of the following: comparison, trend, composition, ranking, distribution, duty cycle, relationship, space; and carrying out data expansion on the input data according to the attribute data and the intention category corresponding to the attribute data to obtain at least one target intention. Therefore, by extracting the attribute data from the input data and combining the intention type of the attribute data, the attribute data can be effectively utilized, and a plurality of target intentions corresponding to the input data are provided for the target object by considering the intention type corresponding to the attribute data, so that the problem of inaccurate intention generation caused by the fact that the generation of the unique intention is only carried out based on the actual input data is avoided.
Description
Technical Field
The present disclosure relates to the field of computer technologies, and in particular, to a data processing method, a data processing device, an electronic device, and a storage medium.
Background
With the development of the internet, various application systems built on the internet are also endless. In order to improve the user experience of an application system, a question and answer function is generally provided for a user in the process of using the application system, and when a user inputs a question, the intention of the user is analyzed through the question input by the user so as to provide an answer matched with the intention of the user. The current intention is usually generated by converting a sentence into a SOL sentence after the application system receives the sentence input by the user, and directly selecting related field data from a data set pre-stored in the application system based on the SQL sentence, wherein the related field data is used as the intention. The current intention generation mode is used for generating the intention only based on the sentence actually input by the user, but sometimes the intention is difficult to describe by the user, and the directions of different users focusing on the same sentence can be different, if the intention generation is also performed based on the actually input sentence, the generated intention of the user is inaccurate, and the user experience is influenced.
Disclosure of Invention
The embodiment of the application provides a data processing method, a data processing device, electronic equipment and a storage medium.
According to a first aspect of the present application, there is provided a data processing method comprising: receiving input data input by a target object; extracting attribute data from the input data to obtain at least one attribute data; determining an intention category corresponding to the attribute data, wherein the intention category comprises at least one of the following: comparison, trend, composition, ranking, distribution, duty cycle, relationship, space; and carrying out data expansion on the input data according to the attribute data and the intention category corresponding to the attribute data to obtain at least one target intention.
According to an embodiment of the present application, the method further comprises: and visually displaying the at least one target intention.
According to an embodiment of the present application, extracting attribute data from the input data includes: extracting attribute data from the input data, including: carrying out natural language analysis on the input data to obtain an analysis result; converting the analysis result into a set statement form to obtain original data; and extracting the attributes of the original data to obtain at least one attribute data, wherein the attribute data comprises attribute words and attribute types.
According to an embodiment of the present application, the attribute types include a nominal type attribute, an ordered type attribute, and a measured type attribute.
According to an embodiment of the present application, the determining the intention category corresponding to the attribute data includes: combining the attribute data with a plurality of preset intention categories to obtain a combined result; determining a plurality of visual combinations meeting visual standards from the combination result; and determining the intention category corresponding to the attribute data according to the plurality of visual combinations.
According to an embodiment of the present application, according to the attribute data and the intention category corresponding to the attribute data, performing data expansion on the input data includes: determining analysis logic corresponding to the attribute data according to a preset map library, the attribute type corresponding to the attribute data and the intention type; and carrying out data expansion on the input data according to the analysis logic of the attribute data.
According to an embodiment of the present application, the preset map library includes attribute maps of a plurality of attribute types, and the attribute maps include a plurality of intention categories and analysis logic corresponding to each intention category.
According to a second aspect of the present application, there is provided a data processing apparatus comprising: the receiving module is used for receiving input data input by a target object; the extraction module is used for extracting attribute data from the input data to obtain at least one attribute data; the determining module is used for determining an intention category corresponding to the attribute data, wherein the intention category comprises at least one of the following: comparison, trend, composition, ranking, distribution, duty cycle, relationship, space; and the expansion module is used for carrying out data expansion on the input data according to the attribute data and the intention category corresponding to the attribute data to obtain at least one target intention.
According to a third aspect of the present application, there is provided an electronic device comprising:
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein,
the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the methods described herein.
According to a fourth aspect of the present application, there is provided a non-transitory computer readable storage medium storing computer instructions for causing the computer to perform the method described herein.
According to the method, input data input by the target object are received; extracting attribute data from the input data to obtain at least one attribute data; determining the intention category corresponding to the attribute data; and carrying out data expansion on the input data according to the attribute data and the intention category corresponding to the attribute data to obtain at least one target intention. Therefore, by extracting the attribute data from the input data and combining the intention type of the attribute data, the attribute data corresponding to the input data can be effectively utilized, and a plurality of target intentions corresponding to the input data are provided for the target object in consideration of the intention type corresponding to the attribute data, so that the problem that the intention generation is inaccurate due to the fact that only the unique intention generation is carried out on the basis of the actual input data is avoided, the plurality of target intentions corresponding to the input data are fully considered, and the accuracy of the intention generation is improved.
It should be understood that the teachings of the present application are not required to achieve all of the above-described benefits, but rather that certain technical solutions may achieve certain technical effects, and that other embodiments of the present application may also achieve benefits not mentioned above.
Drawings
The above, as well as additional purposes, features, and advantages of exemplary embodiments of the present application will become readily apparent from the following detailed description when read in conjunction with the accompanying drawings. Several embodiments of the present application are illustrated by way of example, and not by way of limitation, in the figures of the accompanying drawings, in which:
in the drawings, the same or corresponding reference numerals indicate the same or corresponding parts.
Fig. 1 is a schematic implementation flow diagram of a data processing method according to an embodiment of the present application;
fig. 2 is a schematic implementation flow diagram of a data extraction method of the data processing method according to the embodiment of the present application;
fig. 3 is a schematic implementation flow diagram of an intention category determining method of a data processing method according to an embodiment of the present application;
fig. 4 is a schematic implementation flow diagram of a data expansion method of the data processing method according to the embodiment of the present application;
fig. 5 is a schematic implementation flow diagram of a specific application example of a data processing method according to an embodiment of the present application;
fig. 6 is a schematic diagram showing the composition and structure of a data processing apparatus according to an embodiment of the present application.
Detailed Description
In order to make the objects, features and advantages of the present application more obvious and understandable, the technical solutions of the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is apparent that the described embodiments are only some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments herein without making any inventive effort, are intended to be within the scope of the present application.
Fig. 1 shows a schematic implementation flow chart of a data processing method according to an embodiment of the present application.
Referring to fig. 1, an embodiment of the present application provides an information processing method, including: operation 101, receiving input data input by a target object; operation 102, extracting attribute data from input data to obtain at least one attribute data; operation 103, determining an intention category corresponding to the attribute data, wherein the intention category comprises at least one of the following: comparison, trend, composition, ranking, distribution, duty cycle, relationship, space; and 104, performing data expansion on the input data according to the attribute data and the intention category corresponding to the attribute data to obtain at least one target intention.
In operation 101, input data input by a target object is received.
Specifically, the embodiment of the application can be applied to various application systems with a user questioning function, wherein the application systems include, but are not limited to, an e-commerce system, a financial system, various large platforms and various application software. The target object may include a user of the application system, such as an investor, buyer, etc.
Furthermore, in the process that the target object uses the application system, under the condition that self-service problem consultation is required, input data input by a user can be received first. The input data received from the user may be input data received from a question-answer interaction page provided by the user in the application system.
For example, some investors usually need to know about the invested company through some related platforms or application systems before investing, at this time, the investors can ask questions through the related platforms, for example, input "how like the same flowers are in the red" and then can detect the input data input by the target object, i.e., "how like the same flowers are in the red".
In operation 102, attribute data extraction is performed on the input data to obtain at least one attribute data.
In particular, data may be classified into a plurality of attribute types, for example, a nominal (logical) data attribute (abbreviated as N), an ordered (Ordinal) data attribute (abbreviated as O), and a measured (Interval) data attribute (abbreviated as I). Where N is the most basic attribute type used to represent classification or tag data, there is no order or size relationship between the various classes, e.g., color (red, green, blue), gender (male, female), and country (united states, uk, china) all belong to N. O represents data having a sequential or hierarchical relationship, but cannot accurately measure the magnitude of a difference, e.g., education level (primary, middle, high, college) and product ratings (low, middle, high) all belong to O, where some values are known to be greater or lesser than others, but cannot determine the magnitude of a particular difference. I represents continuous data with equidistant relationships, the magnitude of the difference can be measured, but without absolute zero, e.g., temperature (degrees celsius, degrees fahrenheit) is I, where the differences between the values can be compared and the interval between them calculated.
Further, extracting attribute data from the input data to obtain a plurality of attribute data corresponding to the input data, wherein the attribute data is used for representing a section of data corresponding to each attribute type in the input data. For example, "how the company a is red, the company a represents attribute data corresponding to the Nominal data attribute, and red represents attribute data corresponding to the orinal data attribute.
In an embodiment of the present application, attribute data extraction may be performed by performing data classification on input data, specifically, performing attribute type classification on the input data according to feature information of each attribute type, and classifying attribute data corresponding to each attribute type in the input data.
Specifically, each data attribute type has respective feature information, some feature information that each attribute type has certain exists can be obtained by performing feature information analysis on each attribute type, and then according to a plurality of feature information corresponding to each attribute type, the fields of a plurality of feature information with corresponding attribute types in the input data are determined to be attribute data corresponding to the attribute types.
In an embodiment of the present application, an attribute data extraction model may be trained in advance, and attribute data acquisition may be performed on input data by using the attribute data extraction model. The attribute data extraction model can be obtained by training a common neural network model by taking a large amount of training data as input and taking target attribute data of each training data under each attribute type as output, and the specific training process can refer to a conventional neural network model training process, for example, a convolutional neural network model training process, which is not described herein.
It should be noted that, the extraction method of the attribute data is not particularly limited in this application, and any extraction method capable of extracting the attribute data corresponding to the input data belongs to the protection scope of this application, for example, the attribute data may be extracted by using a named entity recognition model.
Further, although there are a plurality of attribute types, the input data is not associated with the attribute data corresponding to each attribute type, if there is no corresponding field in the input data when extracting a certain attribute type, the default input data is no attribute data of the attribute type, and the existing attribute data is extracted, so that the attribute data extracted by the attribute data is at least one.
In operation 103, an intention category corresponding to the attribute data is determined, the intention category including at least one of: contrast, trend, composition, ranking, distribution, duty cycle, relationship, space.
In an embodiment of the present application, the intent category may be a common visual intent category, specifically including contrast, trend, composition, ranking, distribution, duty cycle, relationship, space, and the like.
The visual intention category corresponding to each attribute data is different, and at least one intention category corresponding to each attribute type can be preconfigured;
after the attribute data is determined, the corresponding intention category is determined according to the attribute type of the attribute data. Wherein the intent category may be considered at least one intent category.
In operation 104, according to the attribute data and the intention category corresponding to the attribute data, data expansion is performed on the input data to obtain at least one target intention.
And respectively carrying out data expansion on each attribute data from the angle of the corresponding intention category to obtain at least one target intention obtained by carrying out data expansion on the input data according to each attribute data.
In one embodiment of the present application, after the at least one target intention is determined, the at least one target intention is visually presented.
Therefore, according to the embodiment of the application, the corresponding attribute data are extracted from the input data, then the intention type corresponding to each attribute data is determined, the input data are expanded from the angles of the attribute data and the intention type, the attribute data of the input data are fully utilized, various intention types possibly involved are considered, the generated intention data are more comprehensive, and the accuracy is higher.
Fig. 2 is a schematic implementation flow diagram of a data extraction method of the data processing method according to the embodiment of the present application.
Referring to fig. 2, in an embodiment of the present application, the step 102 of extracting attribute data from input data includes:
operation 201, performing natural language analysis on input data to obtain an analysis result;
operation 202, converting the analysis result into a set statement form to obtain original data;
and 203, extracting the attributes of the original data to obtain at least one attribute data, wherein the attribute data comprises an attribute word and an attribute type.
In operations 201 to 202, input data is first converted into a set statement form that can be processed by a computer, so as to obtain original data, where the set statement form may be an SQL statement. The specific conversion method can refer to a conventional SQL statement conversion method, and is not described herein.
In operation 203, the raw data is identified by a pre-trained named entity model (Named Entity Recognition, NER), resulting in attribute data comprising attribute words and attribute types. The NER model may be trained to extract attribute words corresponding to a plurality of attribute types in the original data, where the training process of the NER model may refer to a conventional training process of the NER model, which is not described herein.
For example, the input data is "same flower in the same direction as the red", the input data is converted into the form of SOL sentence and then input into the NER model, the NER model can output attribute data 1 with attribute type N and attribute word same flower in the same direction as the red; and attribute type 1, attribute word is attribute data 2 of the segmentation.
Fig. 3 is a schematic implementation flow diagram of an intention category determining method of a data processing method according to an embodiment of the present application.
Referring to fig. 3, in an embodiment of the present application, operation 103, determining an intention category corresponding to attribute data includes:
in operation 301, the attribute data is combined with a plurality of preset intention categories to obtain a combined result.
And combining each attribute data with each preset intention category to obtain a plurality of combinations, namely a combination result.
For example, in the case where the attribute data is 3, 24 combinations, that is, the combination result, are obtained by combining the attribute data (3) and the preset intention category (8).
At operation 302, a plurality of visualization combinations that meet the visualization criteria are determined from the combined results.
After all the combinations are obtained, it is required to determine that the combinations are visualizable, i.e. visualization criteria, so as to determine a plurality of visualizable combinations in the combination result. The visualization standard may be configured according to practical situations, and the application is not specifically limited.
In an embodiment of the present application, when determining the visualization combination, in addition to determining whether the visualization criteria are met, additional conditions may be added to determine the best intention category corresponding to each attribute data. For example, additional conditions other than whether visualization is possible may be added to the visualization standard, for example, taking the combination with the highest similarity as the best combination corresponding to each attribute data, and the best combination is the visualization combination corresponding to the current attribute data.
And operation 303, determining the intention category corresponding to the attribute data according to the plurality of visual combinations.
Each visual combination shows one attribute data and one intention category, the visual combination corresponding to each attribute data is determined, and the intention category is determined from the corresponding visual combination, namely the intention category of each attribute data is obtained. Wherein the intention category of each attribute data is at least one.
The operations 301-303 may be implemented by a support vector machine trained in advance, and the training process of the support vector machine may include: acquiring a training set, wherein the training set comprises a plurality of training combination data and training results corresponding to each training combination data, the training combination data comprises training attribute words and training intention categories, and the training results comprise whether the corresponding training combination data can be visualized or not; and training the support vector machine through a training set, evaluating the support vector machine, and determining that the support vector machine training is completed under the condition that the error of the support vector machine meets the requirement.
Fig. 4 is a schematic implementation flow diagram of a data expansion method of the data processing method according to the embodiment of the present application.
Referring to fig. 4, in an embodiment of the present application, the operation 104 performs data expansion on the input data according to the attribute data and the intention category corresponding to the attribute data, including:
in operation 401, according to a preset map library, an attribute type corresponding to the attribute data, and an intention type, determining analysis logic corresponding to the attribute data.
In an embodiment of the present application, the preset map library includes attribute maps of a plurality of attribute types, and the attribute maps include a plurality of intention categories and analysis logic corresponding to each intention category.
For different attribute types, different attribute maps can be constructed in advance, and the process of constructing the attribute maps can comprise the following steps: acquiring training data, which may be a large amount of unstructured or semi-structured data, such as text documents, web page content, log files, etc.; preprocessing training data, such as text cleaning, word segmentation and the like; identifying attribute words in the preprocessed training data by using the NER model, for example, attribute words corresponding to a certain attribute type; determining association relations among a plurality of attribute words, wherein the association relations can be obtained through a common relation extraction model and are not described in detail herein; according to the association relation between the attribute words, the attribute words are taken as nodes, the association relation is taken as an edge, and an attribute map of the attribute type corresponding to the attribute words, which is also called an attribute relation map, such as an I corresponding index relation map, is constructed. The association relationship may refer to a relationship between two attribute words, for example, an industry, a competitor, association, comparison, and the like, which are not limited in this application.
And then training the association relationship in the attribute map corresponding to each attribute type and the relationship of the intention category through the large language model in advance, and enabling the large language model to have the function of inputting the intention category through training to obtain the association relationship.
And inputting the intention category corresponding to each attribute data into the large model through the large language model trained in advance, so that the association relationship under the corresponding attribute type can be obtained. Wherein the association relationship is analysis logic.
Operation 402, performing data expansion on the input data according to the analysis logic of the attribute data.
After determining the analysis logic of each attribute data, expanding the corresponding attribute words in the input data according to the analysis logic. For example, assuming that the analysis logic is the industry, the input data is "net profit of company a", and the attribute word of the currently expanded attribute data is company a, the input data may be expanded to "net profit of company a and company B and company C of the industry to which the input data belongs".
In order to further understand the technical solutions of the embodiments of the present application, a specific application example is described below.
Fig. 5 is a schematic implementation flow chart of a specific application example of the data processing method according to the embodiment of the present application.
Referring to fig. 5, this specific application example of the present application includes:
s1, receiving a user question.
Specifically, a user question input by a user on a question-answer interaction page provided by an application system is received, and the user question is input data. Illustratively, the user question may be "company A net profit".
S2, determining at least one attribute type corresponding to the user question and an attribute word corresponding to each attribute type according to the set attribute types.
Specifically, the attribute type includes I, N, O, and the attribute word corresponding to I, N, O in the user question is extracted. Illustratively, taking the user question as "net profit of company a" as an example, the corresponding attribute word of N is "company a", and the corresponding attribute word of I is "net profit". The manner in which the attribute terms are obtained may refer to the description of operation 102 in fig. 1, which is not repeated here.
S3, predicting the intention category of each attribute type according to a pre-trained support vector machine to obtain the intention category corresponding to each attribute type.
Specifically, the support vector machine is trained in advance, and the training process of the support vector machine may include: acquiring a training set, wherein each training sample in the training set comprises an attribute type and a visual result of the attribute type and each set intention category, and the visual result comprises a proper presentation and a non-proper presentation; and training the support vector machine through the training set so that the support vector machine outputs the capability of the intention category which corresponds to the attribute type and is suitable for presentation.
The set intention category may be comparison, trend, composition, ranking, distribution, duty ratio, relationship, space.
And inputting the acquired attribute types into a support training machine to obtain the intention category which corresponds to each attribute type and is suitable for presentation (namely, suitable for visualization).
S4, obtaining analysis logic of the intention category corresponding to each attribute type.
Specifically, a knowledge graph corresponding to each attribute type, namely the attribute graph, is constructed in advance, wherein the attribute graph takes an attribute word as a node and takes an association relationship as an edge. After the intention category of each attribute type is obtained, the association relationship corresponding to the intention category of each attribute type is obtained, and the association relationship is analysis logic. An exemplary analysis logic may be another company of the industry, and assuming that the ranking corresponds to another company of the industry in the attribute map corresponding to N, the "other company of the industry to which the analysis logic belongs" may be queried through the "ranking" of the intention category.
And S5, expanding data according to the analysis logic to obtain at least one target intention.
Specifically, the data expansion corresponding to the analysis logic is performed on the attribute words corresponding to each analysis logic, for example, "net profit of company a" is expanded to "net profit of company a and other companies in the industry to which the company belongs", so that at least one target intention is obtained after the data expansion is performed on the attribute words corresponding to each attribute type in the user question.
Fig. 6 is a schematic diagram showing the composition and structure of a data processing apparatus according to an embodiment of the present application.
Referring to fig. 6, based on the above data processing method, an embodiment of the present application further provides a data processing apparatus, where the apparatus includes: a receiving module 501, configured to receive input data input by a target object; the extraction module 502 is configured to perform attribute data extraction on the input data to obtain at least one attribute data; a determining module 503, configured to determine an intention category corresponding to the attribute data, where the intention category includes at least one of: comparison, trend, composition, ranking, distribution, duty cycle, relationship, space; the expansion module 504 is configured to perform data expansion on the input data according to the attribute data and the intention category corresponding to the attribute data, so as to obtain at least one target intention.
In an embodiment of the present application, the apparatus further includes: and the display module is used for visually displaying at least one target intention.
In one embodiment of the present application, the extraction module 502 includes: the analysis sub-module is used for carrying out natural language analysis on the input data to obtain an analysis result; the conversion sub-module is used for converting the analysis result into a set statement form to obtain original data; and the extraction sub-module is used for extracting the attributes of the original data to obtain at least one attribute data, wherein the attribute data comprises an attribute word and an attribute type.
In an embodiment of the present application, the determining module 503 includes: the combination sub-module is used for combining the attribute data with a plurality of preset intention categories to obtain a combination result; a first determining sub-module for determining a plurality of visual combinations meeting the visual standard from the combination result; and the second determining submodule is used for determining the intention category corresponding to the attribute data according to the plurality of visual combinations.
In one embodiment of the present application, the expansion module 504 includes: the third determining module is used for determining analysis logic corresponding to the attribute data according to a preset map library, the attribute type corresponding to the attribute data and the intention type; and the expansion sub-module is used for carrying out data expansion on the input data according to the analysis logic of the attribute data.
It should be noted that, the description of the apparatus in the embodiment of the present application is similar to the description of the embodiment of the method described above, and has similar beneficial effects as the embodiment of the method, so that a detailed description is omitted. The technical details of the data processing apparatus provided in the embodiments of the present application may be understood from the description of any one of fig. 1 to 5.
According to embodiments of the present application, there is also provided an electronic device and a non-transitory computer-readable storage medium.
It should be appreciated that various forms of the flows shown above may be used to reorder, add, or delete steps. For example, the steps described in the present application may be performed in parallel, sequentially, or in a different order, provided that the desired results of the technical solutions disclosed in the present application can be achieved, and are not limited herein.
The foregoing is merely specific embodiments of the present application, but the scope of the present application is not limited thereto, and any person skilled in the art can easily think about changes or substitutions within the technical scope of the present application, and the changes and substitutions are intended to be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (10)
1. A method of data processing, the method comprising:
receiving input data input by a target object;
extracting attribute data from the input data to obtain at least one attribute data;
determining an intention category corresponding to the attribute data, wherein the intention category comprises at least one of the following: comparison, trend, composition, ranking, distribution, duty cycle, relationship, space;
and carrying out data expansion on the input data according to the attribute data and the intention category corresponding to the attribute data to obtain at least one target intention.
2. The method according to claim 1, wherein the method further comprises:
and visually displaying the at least one target intention.
3. The method of claim 1, wherein extracting attribute data from the input data comprises:
carrying out natural language analysis on the input data to obtain an analysis result;
converting the analysis result into a set statement form to obtain original data;
and extracting the attributes of the original data to obtain at least one attribute data, wherein the attribute data comprises attribute words and attribute types.
4. A method according to claim 3, wherein the attribute types include nominal data attributes, ordered data attributes, and measured data attributes.
5. The method of claim 1, wherein the determining the intent category to which the attribute data corresponds comprises:
combining the attribute data with a plurality of preset intention categories to obtain a combined result;
determining a plurality of visual combinations meeting visual standards from the combination result;
and determining the intention category corresponding to the attribute data according to the plurality of visual combinations.
6. The method of claim 1, wherein the data augmentation of the input data according to the attribute data and the intent category to which the attribute data corresponds comprises:
determining analysis logic corresponding to the attribute data according to a preset map library, the attribute type corresponding to the attribute data and the intention type;
and carrying out data expansion on the input data according to the analysis logic of the attribute data.
7. The method of claim 6, wherein the preset atlas library comprises an attribute atlas of a plurality of attribute types, the attribute atlas comprising a plurality of intent categories and analysis logic corresponding to each intent category.
8. A data processing apparatus, the apparatus comprising:
the receiving module is used for receiving input data input by a target object;
the extraction module is used for extracting attribute data from the input data to obtain at least one attribute data;
the determining module is used for determining an intention category corresponding to the attribute data, wherein the intention category comprises at least one of the following: comparison, trend, composition, ranking, distribution, duty cycle, relationship, space;
and the expansion module is used for carrying out data expansion on the input data according to the attribute data and the intention category corresponding to the attribute data to obtain at least one target intention.
9. An electronic device, comprising:
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein,
the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-7.
10. A non-transitory computer readable storage medium storing computer instructions for causing a computer to perform the method of any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410030664.6A CN117540107B (en) | 2024-01-09 | 2024-01-09 | Data processing method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410030664.6A CN117540107B (en) | 2024-01-09 | 2024-01-09 | Data processing method and device, electronic equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117540107A true CN117540107A (en) | 2024-02-09 |
| CN117540107B CN117540107B (en) | 2024-05-07 |
Family
ID=89790394
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410030664.6A Active CN117540107B (en) | 2024-01-09 | 2024-01-09 | Data processing method and device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117540107B (en) |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040064450A1 (en) * | 2002-09-30 | 2004-04-01 | Kabushiki Kaisha Toshiba | Method for preparing data to be analyzed, data analysis method, data analysis device, data preparation program, data analysis program, data prediction device, data prediction method, data prediction program and computer |
| CN107861981A (en) * | 2017-09-28 | 2018-03-30 | 北京奇艺世纪科技有限公司 | A kind of data processing method and device |
| CN113868493A (en) * | 2021-09-13 | 2021-12-31 | 浪潮通信信息系统有限公司 | Visual chart recommendation method |
| CN113946657A (en) * | 2021-10-22 | 2022-01-18 | 唐亮 | Knowledge reasoning-based automatic identification method for power service intention |
| CN114328799A (en) * | 2021-11-10 | 2022-04-12 | 腾讯科技（深圳）有限公司 | Data processing method, device and computer readable storage medium |
| WO2022212563A1 (en) * | 2021-03-30 | 2022-10-06 | Tableau Software, LLC | Applying a visual analytics intent language to generate data visualizations |
| US20220405314A1 (en) * | 2021-06-22 | 2022-12-22 | Adobe Inc. | Facilitating generation of data visualizations via natural language processing |
| CN115827612A (en) * | 2022-11-25 | 2023-03-21 | 中电金信软件有限公司 | Data attribute generation method and device and electronic equipment |
| CN116167355A (en) * | 2021-11-25 | 2023-05-26 | 中移（杭州）信息技术有限公司 | Intention recognition method, device, equipment and storage medium |
| CN116244412A (en) * | 2022-12-20 | 2023-06-09 | 北京航天云路有限公司 | Multi-intention recognition method and device |
-
2024
- 2024-01-09 CN CN202410030664.6A patent/CN117540107B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040064450A1 (en) * | 2002-09-30 | 2004-04-01 | Kabushiki Kaisha Toshiba | Method for preparing data to be analyzed, data analysis method, data analysis device, data preparation program, data analysis program, data prediction device, data prediction method, data prediction program and computer |
| CN107861981A (en) * | 2017-09-28 | 2018-03-30 | 北京奇艺世纪科技有限公司 | A kind of data processing method and device |
| WO2022212563A1 (en) * | 2021-03-30 | 2022-10-06 | Tableau Software, LLC | Applying a visual analytics intent language to generate data visualizations |
| US20220405314A1 (en) * | 2021-06-22 | 2022-12-22 | Adobe Inc. | Facilitating generation of data visualizations via natural language processing |
| CN113868493A (en) * | 2021-09-13 | 2021-12-31 | 浪潮通信信息系统有限公司 | Visual chart recommendation method |
| CN113946657A (en) * | 2021-10-22 | 2022-01-18 | 唐亮 | Knowledge reasoning-based automatic identification method for power service intention |
| CN114328799A (en) * | 2021-11-10 | 2022-04-12 | 腾讯科技（深圳）有限公司 | Data processing method, device and computer readable storage medium |
| CN116167355A (en) * | 2021-11-25 | 2023-05-26 | 中移（杭州）信息技术有限公司 | Intention recognition method, device, equipment and storage medium |
| CN115827612A (en) * | 2022-11-25 | 2023-03-21 | 中电金信软件有限公司 | Data attribute generation method and device and electronic equipment |
| CN116244412A (en) * | 2022-12-20 | 2023-06-09 | 北京航天云路有限公司 | Multi-intention recognition method and device |
Non-Patent Citations (1)
| Title |
|---|
| 贺金龙;付立军;姚郑;吕鹏飞;黄徐胜;: "基于网格LSTM混合算法的地质领域用户意图识别", 计算机系统应用, no. 10, 13 October 2020 (2020-10-13), pages 44 - 52 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117540107B (en) | 2024-05-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106708966B (en) | Junk comment detection method based on similarity calculation | |
| Kim et al. | Using the Bollen-Stine bootstrapping method for evaluating approximate fit indices | |
| Singla et al. | Statistical and sentiment analysis of consumer product reviews | |
| US7689557B2 (en) | System and method of textual information analytics | |
| Hussain et al. | Approximation of COSMIC functional size to support early effort estimation in Agile | |
| US20110153539A1 (en) | Identifying common data objects representing solutions to a problem in different disciplines | |
| US20060200342A1 (en) | System for processing sentiment-bearing text | |
| CN105095444A (en) | Information acquisition method and device | |
| Zhao et al. | Recognition of building group patterns using graph convolutional network | |
| CN110781315A (en) | Food safety knowledge map and construction method of related intelligent question-answering system | |
| Kiefer | Assessing the Quality of Unstructured Data: An Initial Overview. | |
| US20190171713A1 (en) | Semantic parsing method and apparatus | |
| CN106778878B (en) | Character relation classification method and device | |
| Ertz et al. | Review of a proposed methodology for bibliometric and visualization analyses for organizations: application to the collaboration economy | |
| CN111680506A (en) | External key mapping method and device of database table, electronic equipment and storage medium | |
| CN110968664A (en) | Document retrieval method, device, equipment and medium | |
| Rosa et al. | A visual approach for identification and annotation of business process elements in process descriptions | |
| CN111966836A (en) | Knowledge graph vector representation method and device, computer equipment and storage medium | |
| CN117540107B (en) | Data processing method and device, electronic equipment and storage medium | |
| CN111723302A (en) | Recommendation method based on collaborative dual-model deep representation learning | |
| Pajankar et al. | Preparing data for machine learning | |
| US20230126022A1 (en) | Automatically determining table locations and table cell types | |
| Wijesinghe et al. | Computer representation of Venn and Euler diagrams | |
| CN113592512A (en) | Online commodity identity uniqueness identification and confirmation system | |
| WO2016013157A1 (en) | Text processing system, text processing method, and text processing program |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant |