CN114742032A

CN114742032A - Interactive data analysis method, apparatus, device, medium, and program product

Info

Publication number: CN114742032A
Application number: CN202210455201.5A
Authority: CN
Inventors: 宋亮
Original assignee: Guangzhou Yaxin Technology Co ltd
Current assignee: Guangzhou Yaxin Technology Co ltd
Priority date: 2022-04-24
Filing date: 2022-04-24
Publication date: 2022-07-12

Abstract

The embodiment of the application provides an interactive data analysis method, an interactive data analysis device, a related device or equipment, and belongs to the technical field of data analysis. The method comprises the following steps: after receiving a sentence to be analyzed expressed according to a natural language, processing the sentence to be analyzed according to an analysis tool with a Chinese recognition function, obtaining a target keyword of the sentence to be analyzed and a sentence pattern structure formed by part-of-speech marks of the target keyword, obtaining target data according to the target keyword, and displaying the target data to a matched chart set through the sentence pattern structure. The method provided by the application not only combines artificial intelligence and visualization means to perform data analysis, but also effectively reduces the threshold of data analysis.

Description

Interactive data analysis method, apparatus, device, medium, and program product

Technical Field

The present application relates to the field of data analysis, and in particular, to an interactive data analysis method, apparatus, electronic device, computer-readable storage medium, and computer program product.

Background

With the global acceleration of digital economy and the rapid development of 5G, artificial intelligence, internet of things and other related technologies, data has gained general acceptance from the viewpoint of influencing the key strategic resource status of business competition. Only if more data resources are acquired and mastered and the data value is exerted can the data resources occupy the dominant position in a new round of global business competition. The strategic significance of the big data era lies not only in mastering huge data information, but also in discovering and understanding information content and the relationship between information and information. Therefore, data analysis becomes a key step in mining the value of data. Currently, in the field of data analysis, data analysis services are provided for users mainly by providing dashboards and graphical data analysis. For example, a user needs to perform a dragging operation according to a certain logic to implement further self-service analysis, and obviously, a certain threshold exists in the existing analysis means.

In addition, some foreign manufacturers have started to gradually introduce machine learning algorithms and models into the data analysis process, and gradually form the prototype of intelligent data analysis, but the interactive language of semantic recognition and semantic analysis is mainly english, and is relatively in the market blank stage for intelligent conversational intelligent analysis of the chinese language.

Therefore, how to provide a low-threshold data analysis means from the Chinese point of view is a problem that needs to be solved urgently at present.

Disclosure of Invention

The scheme provided by the embodiment of the application aims to solve one of the problems.

According to an aspect of an embodiment of the present application, there is provided an interactive data analysis method, including:

receiving an input sentence to be analyzed, wherein the sentence to be analyzed is a sentence expressed according to a natural language; processing the sentence to be analyzed according to a preset analysis tool with a Chinese recognition function to obtain a sentence pattern structure of the sentence to be analyzed and at least one target keyword, wherein the sentence pattern structure comprises a part-of-speech identifier of each target keyword; and acquiring target data according to at least one target keyword, and displaying the target data through a chart set matched with the sentence pattern structure.

In a possible implementation manner, processing the sentence to be analyzed according to a preset analysis tool configured with a chinese recognition function specifically includes:

calling an analysis tool to perform word segmentation processing on a sentence to be analyzed to obtain at least one keyword to be processed; identifying the part of speech of each keyword to be processed and configuring corresponding part of speech identification; and processing the corresponding keywords to be processed according to the part-of-speech identifier of each keyword to be processed to obtain at least one target keyword and a sentence pattern structure constructed by the part-of-speech identifier carried by each target keyword.

In another possible implementation manner, processing the corresponding keyword to be processed according to the part-of-speech identifier of each keyword to be processed specifically includes:

the following processing is sequentially carried out for each keyword to be processed: if the keyword to be processed does not accord with the preset grammar rule, carrying out regular expression processing on the keyword to be processed, and replacing the keyword to be processed with a processing result; if the part-of-speech identifier of the keyword to be processed is a similar identifier, acquiring a standard word similar to the expression of the keyword to be processed, and replacing the keyword to be processed with the standard word similar to the expression; if the part-of-speech identifier of the keyword to be processed is an unidentified identifier, displaying a replacement entry of the keyword to be processed, responding to an input standard word, and replacing the keyword to be processed with the input standard word; and if the part-of-speech identifier of the keyword to be processed is the time identifier, acquiring time range information corresponding to the keyword to be processed, and replacing the keyword to be processed with the time range information.

In yet another possible implementation manner, before obtaining the target data according to the at least one target keyword, the method further includes:

acquiring a history sentence according to at least one target keyword, wherein the history sentence is related to the at least one target keyword; displaying an entry of a chart corresponding to the historical statement; the entry of the graph corresponding to the historical statement is used for receiving input triggering operation so as to display the graph set corresponding to the historical statement.

determining a target data model from pre-stored data models according to at least one target keyword, and determining a query statement based on the target data model; displaying at least one target keyword and a data query inlet; the data query entry is used for receiving input trigger operation so as to query target data from a table corresponding to the target data model according to a query statement.

In another possible implementation manner, determining a target data model from pre-stored data models according to at least one target keyword specifically includes:

determining keywords meeting conditions according to the part-of-speech identifier of each target keyword, wherein the meeting conditions comprise that the part-of-speech identifier is an identifier for describing dimensions or an identifier for an index; searching matched data models in pre-stored data models according to the keywords meeting the conditions, wherein each pre-stored data model comprises attributes describing dimensions or indexes; and if the number of the matched data models is larger than zero, determining a target data model from the matched data models.

In yet another possible implementation, the process of determining the set of charts that match the sentence pattern structure includes:

matching operation is carried out in a historical instrument board according to the part-of-speech identification in the sentence pattern structure, and the historical instrument board is configured with dimension and index information; if the matching fails, determining a matched chart set from a pre-stored instrument board according to the sentence pattern structure, wherein the pre-stored instrument board is composed of a plurality of charts according to a pre-stored display mode; and if the matching is successful, determining the matched historical dashboard as the matched chart set.

In yet another possible implementation, the target data is presented by a chart matching with the sentence pattern structure, including:

configuring a non-data area of each chart in the instrument panel, wherein the non-data area comprises title information and coordinate axis description information of the corresponding chart; and loading the target data into the instrument panel and displaying the target data.

In yet another possible implementation manner, the method further includes:

and establishing and storing the incidence relation between the statement to be analyzed, the target data and the matched chart so as to facilitate secondary analysis.

According to another aspect of embodiments of the present application, there is provided an interactive data analysis apparatus, including:

the receiving and sending module is used for receiving input sentences to be analyzed, wherein the sentences to be analyzed are sentences expressed according to natural language; the first processing module is used for processing the sentence to be analyzed according to a preset analysis tool with a Chinese recognition function, so as to obtain a sentence pattern structure of the sentence to be analyzed and at least one target keyword, wherein the sentence pattern structure comprises a part-of-speech identifier of each target keyword; the acquisition module is used for acquiring target data according to at least one target keyword; and the display module is used for displaying the target data through the chart set matched with the sentence pattern structure.

According to another aspect of embodiments of the present application, there is provided an electronic device including: memory, a processor and a computer program stored on the memory, the processor executing the computer program to perform the steps of the method according to one aspect described above.

According to yet another aspect of embodiments of the present application, there is provided a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of the method of the one aspect described above.

According to an aspect of embodiments of the present application, there is provided a computer program product comprising a computer program which, when executed by a processor, performs the steps of the method according to the above aspect.

The technical scheme provided by the embodiment of the application has the following beneficial effects:

the embodiment of the application provides an interactive data analysis method, which comprises the following steps: after receiving a sentence to be analyzed expressed according to a natural language, processing the sentence to be analyzed according to an analysis tool with a Chinese recognition function, obtaining a target keyword of the sentence to be analyzed and a sentence pattern structure formed by part-of-speech identification of the target keyword, acquiring target data according to the target keyword, and displaying the target data to a matched chart set through the sentence pattern structure. By receiving input sentences and displaying target processing obtained by processing, an interactive data processing mode is realized, and even ordinary users without special skills can easily realize data analysis; in the processing process of processing the sentences to be analyzed, the sentences to be analyzed are processed by combining the preset identification tools representing intelligent means such as artificial intelligence, and the like, so that the aim of fusing the means such as artificial intelligence and the like and the visual data analysis means is fulfilled. The method provided by the embodiment of the application not only realizes the data analysis by combining artificial intelligence and visualization means, but also effectively reduces the threshold of the data analysis.

Drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings used in the description of the embodiments of the present application will be briefly described below.

Fig. 1 is a schematic diagram of a system architecture for implementing interactive data analysis according to an embodiment of the present application;

fig. 2 is a schematic flowchart of an interactive data analysis method according to an embodiment of the present application;

fig. 3a is a schematic structural diagram of an interactive data analysis apparatus according to an embodiment of the present application;

FIG. 3b is a schematic structural diagram of another interactive data analysis apparatus according to an embodiment of the present application;

fig. 4 is a schematic structural diagram of an electronic device according to an embodiment of the present application.

Detailed Description

Embodiments of the present application are described below in conjunction with the drawings in the present application. It should be understood that the embodiments set forth below in connection with the drawings are exemplary descriptions for explaining technical solutions of the embodiments of the present application, and do not limit the technical solutions of the embodiments of the present application.

As used herein, the singular forms "a", "an", "the" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should be further understood that the terms "comprises" and/or "comprising," when used in this specification in connection with embodiments of the present application, specify the presence of stated features, information, data, steps, operations, elements, and/or components, but do not preclude the presence or addition of other features, information, data, steps, operations, elements, components, and/or groups thereof, as embodied in the art. It will be understood that when an element is referred to as being "connected" or "coupled" to another element, it can be directly connected or coupled to the other element or intervening elements may be present. Further, "connected" or "coupled" as used herein may include wirelessly connected or wirelessly coupled. The term "and/or" as used herein indicates at least one of the items defined by the term, e.g., "a and/or B" may be implemented as "a", or as "B", or as "a and B".

To make the objects, technical solutions and advantages of the present application more clear, the following detailed description of the embodiments of the present application will be made with reference to the accompanying drawings.

The following description of the terms and related art to which this application relates:

data visualization technology: data visualization refers to a technique for visually expressing data interaction using the perception ability of the human eye to enhance cognition. Data visualization generally includes scientific visualization, information visualization, and visualization analysis 3 classes. Scientific visualization mainly realizes specific data visualization, and focuses on data with natural geometric structures, such as magnetic fields, geographic structures and the like; information visualization focuses on visualization of abstract data, such as a tree diagram and a histogram; visualization analysis refers to the combination of data mining and other knowledge in data visualization, such as analytical reasoning, visual presentation, interaction and the like.

And (3) enhanced data analysis: the method mainly refers to a method and a process for completing complex data analysis by simple natural language dialogue under the condition that a common user is helped by a simple natural language dialogue through depositing common and universal data analysis scenes into a product by applying technologies such as machine learning and artificial intelligence based data analysis and BI functions and without assistance of data science experts or IT personnel.

NLP (Natural LanguageProcess) technique: natural language processing includes natural language understanding, which is the conversion of natural language into a computer-understandable language or the conversion of unstructured text into structured information, and natural language generation. The language is a sound (image) instruction with a unified coding/decoding standard, which is made by the communication requirements among the similar organisms, and comprises body languages such as gestures, voice and the like, and characters are imaging symbols; natural language generally refers to a language that naturally evolves with culture, such as chinese, english, japanese, etc.; natural languages are distinguished from man-made languages such as world languages, programming languages, etc.

NL2SQL technique (Natural Language to SQL): the technology for converting natural language into SQL sentences can serve as an intelligent interface of the database, so that users who are not familiar with the database can quickly find the data required by the users. For example, if a user enters a statement or several keywords in a search box, the user immediately can obtain an SQL statement or query result.

For more than twenty years, the main data analysis mode is visual instrument panel and report, the threshold of data analysis is high, and certain time and professional knowledge are needed for mastering the data analysis, so that the data value cannot be popularized rapidly. With the rapid development of technologies such as artificial intelligence, knowledge maps and the like, in the future, a user can even operate in natural language, the user can speak a demand like a conversation with a person, then the system gives required data and even an analysis result, the data analysis threshold is reduced to 'infinite low', and the obtained result has depth. However, this is only one idea.

In terms of the background technology, with the large scale of the data volume and the continuous improvement of the analysis requirement in the future, the data analysis technology will be expanded and enhanced in multiple dimensions. As the amount of data is greatly increased, the existing processing and analyzing technology may not meet the requirement of timely analysis of data, so the intelligent interactive data analysis capability will be an important field for the development of future data analysis technology.

In order to solve at least one technical problem existing in the background art or to improve the technical problem, the present application provides an interactive data analysis scheme, which reduces the data analysis threshold by an interactive question-answer processing manner.

The technical solutions of the embodiments of the present application and the technical effects produced by the technical solutions of the present application will be described below through descriptions of several exemplary embodiments. It should be noted that the following embodiments may be referred to, referred to or combined with each other, and the description of the same terms, similar features, similar implementation steps and the like in different embodiments is not repeated.

Referring to fig. 1, an embodiment of the present application provides a schematic diagram of a system architecture for implementing interactive data analysis. The system architecture consists of a user-oriented front-end portion, and a back-end portion that provides service support.

Specifically, the front-end portion mainly provides a user-oriented presentation function to guide the user to perform natural language input, and present various types of processed results and the like to the user. For example, the page is presented as follows: a search entry page, a results presentation page, a help page, a recommendation list page, a secondary analysis page, a question and answer page, etc.

Specifically, the back-end portion mainly provides various service modules for analyzing and processing an input sentence. For example, the following services: help services (providing help from a sentence or phrase perspective), word recommendation services (querying and recommending standard words according to any word), user preference services (recording user preferences and combining the preference services for recommendation during querying), dashboard recommendation services (recommending matched dashboards according to processing results), semantic processing services (including regular expression processing, participle processing, near-synonym processing, unmarked word processing, model identification processing, data formatting processing, area identification processing, date conversion processing, and the like), and the like.

It should be noted that the display pages listed in the front end section and the services listed in the back end section are only used as examples to illustrate the application, and should not be construed as limitations of the application. Moreover, those skilled in the art may add appropriate amounts of content to either the front-end portion or the back-end portion, as desired.

The present application further provides a process for implementing interactive data analysis based on the above system architecture, where the process includes steps S110 to S150.

And S110, initializing a question and answer page.

Specifically, the question and answer page comprises an entry of a main interface, and the entry comprises a search box; the question and answer page further includes help information (e.g., help service in the above architecture), a specified data model, and the like, and the specified data model may be a data model created according to information of the user. The information of the user includes: the user's occupation, interests, once search operations, etc.

Wherein, the associated modules or services in this step include: search entry pages, help services.

And S120, questioning interaction.

Specifically, a question sentence or a phrase is input in a question and answer page; through word segmentation processing, keyword prompt and other processing, a plurality of keywords corresponding to input information are obtained, the keywords are matched with standard words prestored in a system, and a part-of-speech identifier is added to each target keyword. And prompting the user for the unrecognizable words (namely, the words with part-of-speech identifiers being unrecognized identifiers) to acquire the standard words input by the user. And finally, determining at least one target keyword and a sentence pattern structure consisting of the part-of-speech identifiers of each target keyword. And calling a machine learning tool configured with the NL2SQL technology to process the target keyword to obtain an SQL statement, and inquiring through the SQL statement to obtain target data.

Wherein, the associated module or service in this step includes: search entry pages, semantic processing services, and the like.

And S130, presenting the result.

Specifically, the dimension and index information are determined according to the processing result in step S220 in the following embodiments of the present application to determine a matching instrument panel. And calling the matched instrument board to display the target data.

Wherein, the associated module or service in this step includes: and a result display page and a dashboard recommendation service.

And S140, issuing a result.

Specifically, the dashboard of the "result presentation" output is saved. After saving, the saved dashboard can be viewed through the data question and answer list, or further edited, or deleted.

And S150, secondary analysis.

Specifically, after the analysis result is obtained, the screening condition may be changed, or the index and the dimension information may be added, and the steps of "question interaction", result presentation, and the like may be repeated, so as to regenerate a new processing result based on the original result. This step may follow, among other things, "result presentation" or "result publishing".

Referring to fig. 2, an interactive data analysis method applied to a terminal is provided in the embodiment of the present application, and the method includes steps S210 to S230.

S210, receiving an input sentence to be analyzed, wherein the sentence to be analyzed is a sentence expressed according to a natural language.

Specifically, the manner of receiving the sentence to be analyzed input by the user may be various, for example: receiving sentences input by a user through a text search box and expressed in natural language through a preset question and answer page, or receiving voice information input by the user through a voice collecting tool. In short, both of the above two modes are operable by an ordinary user.

S220, processing the sentence to be analyzed according to a preset analysis tool with a Chinese recognition function, and obtaining a sentence pattern structure of the sentence to be analyzed and at least one target keyword, wherein the sentence pattern structure comprises a part-of-speech identifier of each target keyword.

And the at least one target keyword is the processed target keyword, and the number of the target keywords is one or more.

And S230, acquiring target data according to the at least one target keyword, and displaying the target data through a chart set matched with the sentence pattern structure.

The embodiment of the application provides an interactive data analysis method, which comprises the following steps: after receiving a sentence to be analyzed expressed according to a natural language, processing the sentence to be analyzed according to a preset Chinese recognition analysis tool to obtain a target keyword of the sentence to be analyzed and a sentence pattern structure formed by part-of-speech marks of the target keyword, acquiring target data according to the target keyword, and displaying the target data to a matched chart set through the sentence pattern structure. By receiving input sentences and displaying the target processing obtained by processing, an interactive data processing mode is realized, and the threshold of data analysis is reduced; in the processing process of processing the sentences to be analyzed, the sentences to be analyzed are processed by combining the preset identification tools representing intelligent means such as artificial intelligence, and the like, so that the aim of fusing the means such as artificial intelligence and the like and the visual data analysis means is fulfilled. The method provided by the application not only realizes the data analysis by combining artificial intelligence and visualization means, but also effectively reduces the threshold of the data analysis.

The embodiment of the present application further provides a possible implementation manner, and S220 may specifically include S221 to S223.

S221, an analysis tool is called to perform word segmentation processing on the sentence to be analyzed, and at least one keyword to be processed is obtained.

Specifically, the sentence to be analyzed is preliminarily split according to a preset semantic rule to obtain at least one keyword to be processed. Wherein, the predetermined analysis tool may be an analysis tool configured with NLP technology. S222, identifying the part of speech of each keyword to be processed and configuring corresponding part of speech identification.

The method for processing the word identity in the terminal comprises the following steps that information of a plurality of word identity identifications is prestored in the terminal, and the method comprises the following steps: time identification, index identification, dimension identification, condition value identification, dimension value identification, first identifiers, sequencing identification, similar identification and unidentified identification.

In one example, there are specific identification means as follows: the time mark is time and is used for describing time, and can be a time point or a range; the index is identified as kpi; dimension identification is dim; the condition value is marked as filter _ value and is used for describing the condition of screening data; the dimension value identifier is filter _ value _ dim, and the specific dimension value is described by combining the dimension identifier; the first few digits are labeled top and are used to describe the first few digits at a certain index angle, e.g., the first 3 income names; the sorting identifier is an order and is used for sorting the data according to a certain rule; similar marks are similar and are used for describing keywords with similar standard words, but the keywords are not the standard words; the unrecognized label is a none, which is used to describe the unrecognized keyword.

In order to more clearly understand the relationship between the part of speech identifier and the keyword to be processed, the embodiment of the present application further provides a plurality of sentence processing examples, which are specifically shown as follows.

Statement 1: 3 in the first income of Jiangsu province in the last half year.

Firstly, performing word segmentation processing on a statement 1 to obtain the following keywords to be processed: "the first half year", "Jiangsu province", "income" of each city ", and" the first 3 cases ".

Secondly, identifying each keyword to be processed, the part-of-speech identifier of each keyword can be determined: the 'first half year' is time; "Jiangsu province" is filter _ value _ dim; "income" is kpi; "commercially available" is dim; the "first 3 cases" is top.

Statement 2: and (4) increasing the income of each province in the last month.

Performing word segmentation processing on the sentence 2 to obtain a plurality of keywords to be processed including the keywords in the previous month and the keywords in the increasing sequence, performing identification processing on the keywords, and determining the part-of-speech identifier as follows: "ascending order" is order, and "last month" is top.

Statement 3: income of provinces in a month is higher than 300000 in prefecture.

After the word segmentation processing is performed on the sentence 3, a plurality of keywords to be processed including "a month" and "above 300000" are obtained, recognition processing is performed on the keywords, and the part-of-speech identifier of each keyword is determined: "month" is none, "above 300000" is filter _ value _ dim.

And S223, processing the corresponding keywords to be processed according to the part-of-speech identifier of each keyword to be processed to obtain the at least one target keyword and a sentence pattern structure constructed by the part-of-speech identifier carried by each target keyword.

Specifically, the following steps a to D are sequentially performed for each keyword to be processed:

A. and if the keyword to be processed does not accord with the preset grammar rule, performing regular expression processing on the keyword to be processed, and replacing the keyword to be processed with a processing result.

Specifically, some keywords to be processed have a problem that the expression does not conform to the grammar rule, so regular expression processing is required to obtain a processing result with a correct expression. For example, after regular expression processing is performed on the keywords such as "above 3000000", "top 3 cases", "prefecture", and the like, the obtained processing result is: "higher than 3000000", "first 3", "prefecture". It should be noted that reference may be made to the prior art regarding the processing of regular expressions.

B. If the part-of-speech identifier of the keyword to be processed is a similar identifier, acquiring a standard word similar to the expression of the keyword to be processed, and replacing the keyword to be processed with the standard word similar to the expression.

Specifically, part of the keywords to be processed cannot be matched with the same keywords in the terminal, and the part of the keywords to be processed needs to be replaced by standard words with similar expressions. For example, the standard word replacement is performed for the above "each city", and the obtained processing result is: "the city of land".

The terminal is pre-stored with a large number of standard words, which can be classified according to the characteristics of industry, field, etc., and continuously absorb new standard words in the process of word segmentation processing.

C. If the part-of-speech identifier of the keyword to be processed is an unidentified identifier, displaying a replacement entry of the keyword to be processed, and replacing the keyword to be processed with an input standard word in response to the input standard word.

Specifically, for the keywords to be processed, the part-of-speech identifiers of which the parts-of-speech cannot be distinguished, are unified as the unidentified identifiers. For the type of the keywords to be processed, the keywords to be processed need to be processed by the user, for example, the keywords to be processed carrying the unidentified identifier are displayed to the user, the user is prompted to replace the keywords, and after the user inputs the terms meeting the conditions, the terms are processed by the keywords to be processed. Wherein, when showing to the user, can provide a plurality of words that can be used for filtering. For example, the user inputs "month 2 last" instead of "month", and the part-of-speech identifier of the new keyword to be processed is also replaced with the time identifier.

D. And if the part-of-speech identifier of the keyword to be processed is the time identifier, acquiring time range information corresponding to the keyword to be processed, and replacing the keyword to be processed with the time range information.

For example, the time range information of the above "first year" is acquired as follows: [ 1/2021/6/30 ].

It should be noted that, after the above processing and replacing operations are sequentially performed on the to-be-processed keywords, the finally determined to-be-processed keywords are determined as the target keywords. And combining the part-of-speech identifiers of all the target keywords together according to the sequence to obtain a sentence pattern structure.

For example, the target keyword in the processing result of statement 1 is (shown in the form of [ keyword, part of speech identifier "): [2021-1-1, 2021-6-30], time ] [ Jiangsu province, filter _ value _ dim ] [ city, dim ] [ income, kpi ] [ first 3 name, top ], and the sentence pattern structure thereof is [ time ] [ filter _ value _ dim ] [ kpi ] [ top ].

In one possible implementation, the obtained target keywords are sequentially displayed.

By displaying the target keywords, valuable information in the input sentence to be analyzed is informed to the user, so that better user experience is provided.

In one example, a question-and-answer interface may be provided, on which statement 1 of the statement to be analyzed and the processing result of statement 1 are presented: [2021-1-1, 2021-6-30] [ Jiangsu province ] [ DINNING City ] [ income ] [ first 3 names ].

In one possible implementation, before obtaining the target data according to the at least one target keyword, the method further includes:

obtaining a history sentence according to the at least one target keyword, wherein the history sentence and the at least one target keyword have correlation; and displaying the entry of the chart corresponding to the historical statement.

Specifically, according to the matching between one or more target keywords in the at least one target keyword and the history keywords pre-stored in the terminal, the history sentence corresponding to the history keyword which is successfully matched is obtained. The terminal stores a processing result of each statement to be analyzed, wherein the processing result comprises the statement to be analyzed, all target keywords of the statement to be analyzed, a chart set loaded with target data and the like.

And the entry of the chart corresponding to the historical statement is used for receiving input triggering operation so as to display the chart set corresponding to the historical statement.

In one example, each history statement may also be presented in the form of a tag on the question-and-answer interface described above. And after any label is clicked, entering a page for displaying the chart set corresponding to the corresponding historical statement.

In a possible implementation manner, before obtaining the target data according to the at least one target keyword, the method may further include:

determining a target data model from the pre-stored data models according to the at least one target keyword, and determining a query statement based on the target data model; and displaying the at least one target keyword and the data query inlet.

In the initialization stage of the terminal, a plurality of pre-stored data models are obtained, wherein the data models are obtained according to data tables stored in a database. For example, the following is performed for each data table: extracting all attributes of the data table, identifying part-of-speech identification of each attribute, gathering the attributes carrying the part-of-speech identification in an array, and taking the array as a data model of the data table. After the data model of each data table is acquired, the data model is stored in the terminal.

The process of determining the target data model may include: determining keywords meeting conditions according to the part-of-speech identifier of each target keyword, wherein the meeting conditions comprise that the part-of-speech identifier is an identifier for describing dimensions or an identifier for indexes; searching matched data models in pre-stored data models according to the keywords meeting the conditions, wherein each pre-stored data model comprises attributes describing dimensions or indexes; and if the number of the matched data models is larger than zero, determining a target data model from the matched data models. Wherein, the number of the keywords meeting the conditions is one or more.

And the terminal also sets a unique code aiming at the standard words of which the part-of-speech identifiers are the dimension identifiers, the index identifiers and the dimension value identifiers. For example, the part of speech of the standard words "male" and "female" is identified as dimension value, which is a specific dimension value of the standard word "gender", and the part of speech of "gender" is identified as dimension. If the code for setting gender is 1000, the codes for "male" and "female" are 1000-1 and 1000-2, respectively, and in the terminal, the code 1000 is associated with the codes 1000-1 and 1000-2.

Specifically, searching for a matched data model in a pre-stored data model according to a keyword meeting a condition may specifically include: and taking all the keywords meeting the conditions as an array, and matching the array with a plurality of pre-stored data models. In the matching process, the attributes with the same part-of-speech identifiers and the keywords are sequentially compared one by one, and whether the attributes and the keywords are associated or not is determined so as to determine whether the attributes and the keywords are corresponding to each other or not. The criterion for judging the association is whether the codes of the two are associated or the same

In one example, the keywords that meet the condition in the processing result of statement 1 include: [ Jiangsu province, filter _ value _ dim ] [ di city, dim ] [ income, kpi ]. Comparing each qualified keyword in the processing result of the statement 1 with a plurality of pre-stored data models, and finding that all attributes in the user information table can cover all the qualified keywords. The data model of the user information table comprises the following attributes: [ NAME, dim ] [ AGE, dim ] [ PROPERTIES, DIM ] [ CAUSEHOLD, DIM ] [ DIM ], DIM ] [ BUTE, DIM ] [ RECEIVING, kpi ].

In one possible implementation, after the target data model is determined, information of a data table corresponding to the target data model is acquired.

In one possible implementation, the process of determining a query statement based on a target data model may include: and calling a preset machine learning algorithm to process the at least one target keyword to obtain a query sentence. The information of the data table may be added in the processing process, or the information of the data table may be added at a corresponding position after the query statement is obtained.

In one example, the preset machine learning algorithm may be a core algorithm of NL2SQL technology. NL2SQL technology, native Language to SQL, converts Natural Language into a technology that can execute SQL statements. NL2SQL is a prior art, and those skilled in the art can refer to the prior art, and therefore, the description thereof is omitted here for simplicity.

In one possible implementation, the data query entry is configured to receive an input trigger operation to query the target data from the table corresponding to the target data model according to a query statement.

In one example, the data query portal may be a search button disposed on a question-and-answer interface. After receiving the triggering operation of the user for the search button, querying the target data from the model according to the determined query statement.

The embodiment of the present application further provides a possible implementation manner, and the process of determining the chart set matched with the sentence pattern structure includes:

and performing matching operation in a historical dashboard according to the part-of-speech identifier in the sentence pattern structure, wherein the historical dashboard is configured with dimension and index information. If the matching fails, determining a matched chart set from a pre-stored instrument board according to the sentence pattern structure, wherein the pre-stored instrument board consists of a plurality of charts according to a pre-stored display mode; and if the matching is successful, determining the matched historical dashboard as the matched chart set.

In the stage of initializing the terminal, a plurality of pre-stored instrument panels are obtained. The pre-stored instrument panels are obtained by training according to a plurality of data tables and charts in a matching display mode in a preset machine training stage. For example, the data table is output to a plurality of charts for displaying according to a plurality of different dimensions and index combinations, a plurality of charts with the best display effect in each group of dimensions and index combinations are screened out and combined together to form an instrument panel, and the dimension and index information of the instrument panel is the information of corresponding indexes and dimension combinations.

Wherein, this instrument board of prestoring can be general instrument board, can most show demands of adaptation.

In one possible implementation, presenting the target data through a chart matching with the sentence pattern structure may include: configuring a non-data area of each chart in the instrument panel, wherein the non-data area comprises title information and coordinate axis description information of the corresponding chart; and loading the target data into the instrument panel and displaying the target data.

Specifically, the non-data area is specific information for describing each chart, and may include other information in addition to the above-described header information and coordinate axis description information. This is not limited by the present application.

The embodiment of the present application further provides a possible implementation manner, and the method further includes:

and issuing the statement to be analyzed, the at least one target keyword and the chart set loaded with the target data.

Specifically, a publication page is generated, and the publication page comprises the statement to be analyzed, the at least one target keyword, and a chart set loaded with target data. The page is then published to the user.

In one possible implementation, the method further includes:

and establishing and storing the incidence relation among the statement to be analyzed, the target data and the matched chart set so as to facilitate secondary analysis.

Specifically, the process of the second analysis includes adding or deleting any target keyword, and then determining a new sentence pattern structure, and new target data. A new dashboard may be determined based on the new sentence pattern structure, and then new target data may be presented. The process may refer to the above embodiments, and is not described herein again for simplicity.

After the information is stored, the statement to be analyzed becomes a history statement, and the matched dashboard also becomes a history dashboard. In the above embodiment, each processing procedure can be put into the next analysis procedure as experience by using the history sentence, the history dashboard, and the like, so that the next processing procedure can be enriched.

Referring to fig. 3, an embodiment of the present application provides an interactive data analysis apparatus, where the apparatus 300 may specifically include: the system comprises a transceiver module 310, a first processing module 320, an acquisition module 330 and a display module 340.

The receiving and sending module 310 is configured to receive an input sentence to be analyzed, where the sentence to be analyzed is a sentence expressed according to a natural language; a first processing module 320, configured to process a sentence to be analyzed according to a preset analysis tool configured with a chinese recognition function, to obtain a sentence pattern structure of the sentence to be analyzed and at least one target keyword, where the sentence pattern structure includes a part-of-speech identifier of each target keyword; an obtaining module 330, configured to obtain target data according to the at least one target keyword; and the display module 340 is used for displaying the target data through the chart set matched with the sentence pattern structure.

In a possible implementation manner, the first processing module 320 is specifically configured to, in processing the sentence to be analyzed according to the preset analysis tool configured with the chinese recognition function to obtain the sentence pattern structure and the at least one target keyword of the sentence to be analyzed:

calling an analysis tool to perform word segmentation processing on a sentence to be analyzed to obtain at least one keyword to be processed; identifying the part of speech of each keyword to be processed and configuring corresponding part of speech identification; and processing the corresponding keywords to be processed according to the part-of-speech identifiers of the keywords to be processed to obtain the at least one target keyword and a sentence pattern structure constructed by the part-of-speech identifiers carried by the target keywords.

In a possible implementation manner, the first processing module 320, in processing the corresponding keyword to be processed according to the part-of-speech identifier of each keyword to be processed, is specifically configured to:

In a possible implementation manner, the apparatus 300 further includes a second processing module 350, and before the second processing module 350 acquires the target data according to the at least one target keyword, the second processing module is specifically configured to:

obtaining a history sentence according to the at least one target keyword, wherein the history sentence has correlation with the at least one target keyword; displaying an entry of a chart corresponding to the historical statement; and the entry of the chart corresponding to the historical statement is used for receiving input triggering operation so as to display the chart set corresponding to the historical statement.

In a possible implementation manner, the apparatus 300 further includes a third processing module 360, before the third processing module 360 obtains the target data according to the at least one target keyword, specifically configured to:

And the data query entry is used for receiving input trigger operation so as to query the target data from the table corresponding to the target data model according to the query statement.

In a possible implementation manner, the third processing module 360, in determining the target data model from the pre-stored data models according to the at least one target keyword, is specifically configured to:

In a possible implementation manner, the apparatus further includes a fourth processing module 370, where the fourth processing module 370, in the process of determining the chart set matching with the sentence pattern structure, is specifically configured to:

In one possible implementation, the presentation module 340 is specifically configured to, in presenting the target data by a chart matching with the sentence pattern structure:

configuring a non-data area of each chart in the instrument panel, wherein the non-data area comprises the title information and coordinate axis description information of the corresponding chart; and loading the target data into the instrument panel and displaying the target data.

In a possible implementation manner, the apparatus 300 further includes a fifth processing module 380, where the fifth processing module 380 is specifically configured to:

An embodiment of the present application provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory, where the processor executes the computer program to implement the steps of the method shown in the above embodiment.

In an alternative embodiment, an electronic device is provided, as shown in fig. 4, the electronic device 4000 shown in fig. 4 comprising: a processor 4001 and a memory 4003. Processor 4001 is coupled to memory 4003, such as via bus 4002. Optionally, the electronic device 4000 may further include a transceiver 4004, and the transceiver 4004 may be used for data interaction between the electronic device and other electronic devices, such as transmission of data and/or reception of data. In addition, the transceiver 4004 is not limited to one in practical applications, and the structure of the electronic device 4000 is not limited to the embodiment of the present application.

The Processor 4001 may be a CPU (Central Processing Unit), a general-purpose Processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array) or other Programmable logic device, a transistor logic device, a hardware component, or any combination thereof. Which may implement or perform the various illustrative logical blocks, modules, and circuits described in connection with the disclosure. The processor 4001 may also be a combination that performs a computational function, including, for example, a combination of one or more microprocessors, a combination of a DSP and a microprocessor, or the like.

Bus 4002 may include a path that carries information between the aforementioned components. The bus 4002 may be a PCI (Peripheral Component Interconnect) bus, an EISA (Extended Industry Standard Architecture) bus, or the like. The bus 4002 may be divided into an address bus, a data bus, a control bus, and the like. For ease of illustration, only one thick line is shown in FIG. 4, but this does not indicate only one bus or one type of bus.

The Memory 4003 may be a ROM (Read Only Memory) or other types of static storage devices that can store static information and instructions, a RAM (Random Access Memory) or other types of dynamic storage devices that can store information and instructions, an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical Disc storage, optical Disc storage (including Compact Disc, laser Disc, optical Disc, digital versatile Disc, blu-ray Disc, etc.), a magnetic Disc storage medium, other magnetic storage devices, or any other medium that can be used to carry or store a computer program and that can be Read by a computer, without limitation.

The memory 4003 is used for storing computer programs for executing the embodiments of the present application, and is controlled by the processor 4001 to execute. The processor 4001 is used to execute computer programs stored in the memory 4003 to implement the steps shown in the foregoing method embodiments.

Among them, electronic devices include but are not limited to: and a server.

Embodiments of the present application provide a computer-readable storage medium, on which a computer program is stored, and when being executed by a processor, the computer program may implement the steps and corresponding contents of the foregoing method embodiments.

Embodiments of the present application further provide a computer program product, which includes a computer program, and when the computer program is executed by a processor, the steps and corresponding contents of the foregoing method embodiments can be implemented.

The terms "first," "second," "third," "fourth," "1," "2," and the like in the description and in the claims of the present application and in the above-described drawings (if any) are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It should be understood that the data so used are interchangeable under appropriate circumstances such that the embodiments of the application described herein are capable of operation in other sequences than illustrated or otherwise described herein.

It should be understood that, although each operation step is indicated by an arrow in the flowchart of the embodiment of the present application, the implementation order of the steps is not limited to the order indicated by the arrow. In some implementation scenarios of the embodiments of the present application, the implementation steps in the flowcharts may be performed in other sequences as desired, unless explicitly stated otherwise herein. In addition, some or all of the steps in each flowchart may include multiple sub-steps or multiple stages based on an actual implementation scenario. Some or all of these sub-steps or stages may be performed at the same time, or each of these sub-steps or stages may be performed at different times, respectively. In a scenario where execution times are different, an execution sequence of the sub-steps or the phases may be flexibly configured according to requirements, which is not limited in the embodiment of the present application.

The foregoing is only an optional implementation manner of a part of implementation scenarios in this application, and it should be noted that, for those skilled in the art, other similar implementation means based on the technical idea of this application are also within the protection scope of the embodiments of this application without departing from the technical idea of this application.

Claims

1. A method of interactive data analysis, the method comprising:

receiving an input sentence to be analyzed, wherein the sentence to be analyzed is a sentence expressed according to a natural language;

processing the sentence to be analyzed according to a preset analysis tool with a Chinese recognition function to obtain a sentence pattern structure of the sentence to be analyzed and at least one target keyword, wherein the sentence pattern structure comprises a part-of-speech identifier of each target keyword;

and acquiring target data according to the at least one target keyword, and displaying the target data through a chart set matched with the sentence pattern structure.

2. The method according to claim 1, wherein the processing the sentence to be analyzed according to the analysis tool configuring the chinese recognition function in advance comprises:

calling the analysis tool to perform word segmentation processing on the sentence to be analyzed to obtain at least one keyword to be processed; identifying the part of speech of each keyword to be processed and configuring corresponding part of speech identification;

and processing the corresponding keywords to be processed according to the part-of-speech identifier of each keyword to be processed to obtain the at least one target keyword and the sentence pattern structure constructed by the part-of-speech identifier carried by each target keyword.

3. The method according to claim 2, wherein the processing the corresponding keyword to be processed according to the part-of-speech identifier of each keyword to be processed comprises:

the following processing is sequentially carried out for each keyword to be processed:

if the keyword to be processed does not accord with the preset grammar rule, carrying out regular expression processing on the keyword to be processed, and replacing the keyword to be processed with a processing result;

if the part-of-speech identifier of the keyword to be processed is a similar identifier, acquiring a standard word similar to the expression of the keyword to be processed, and replacing the keyword to be processed with the standard word similar to the expression;

if the part-of-speech identifier of the keyword to be processed is an unidentified identifier, displaying a replacement entry of the keyword to be processed, responding to an input standard word, and replacing the keyword to be processed with the input standard word;

and if the part-of-speech identifier of the keyword to be processed is the time identifier, acquiring time range information corresponding to the keyword to be processed, and replacing the keyword to be processed with the time range information.

4. The method of claim 1, wherein prior to obtaining target data based on the at least one target keyword, the method further comprises:

obtaining a history statement according to the at least one target keyword, wherein the history statement and the at least one target keyword have correlation;

displaying an entry of a chart corresponding to the historical statement;

5. The method of claim 1, wherein prior to obtaining target data based on the at least one target keyword, the method further comprises:

determining a target data model from pre-stored data models according to the at least one target keyword, and determining a query statement based on the target data model;

displaying the at least one target keyword and a data query inlet;

6. The method of claim 5, wherein determining a target data model from pre-stored data models based on the at least one target keyword comprises:

determining keywords meeting conditions according to the part-of-speech identifier of each target keyword, wherein the meeting conditions comprise that the part-of-speech identifier is an identifier describing dimensions or an identifier of an index;

searching matched data models in the pre-stored data models according to the keywords meeting the conditions, wherein each pre-stored data model comprises attributes describing dimensions or indexes;

and if the number of the matched data models is larger than zero, determining the target data model from the matched data models.

7. The method of claim 1, wherein determining the set of charts that match the sentence structure comprises:

matching operation is carried out in a historical instrument board according to the part-of-speech identification in the sentence pattern structure, and the historical instrument board is configured with dimension and index information;

if the matching fails, determining the matched chart set from a pre-stored instrument board according to the sentence pattern structure, wherein the pre-stored instrument board is composed of a plurality of charts according to a pre-stored display mode;

and if the matching is successful, determining the matched historical instrument board as the matched chart set.

8. The method of claim 7, wherein said presenting said object data by a graph matching said sentence pattern structure comprises:

configuring a non-data area of each chart in the instrument panel, wherein the non-data area comprises title information and coordinate axis description information of the corresponding chart;

and loading the target data into the instrument panel and displaying.

9. The method of claim 1, further comprising:

10. An interactive data analysis apparatus, the apparatus comprising:

the system comprises a receiving and sending module, a processing module and a processing module, wherein the receiving and sending module is used for receiving input sentences to be analyzed, and the sentences to be analyzed are sentences expressed according to natural language;

the first processing module is used for processing the sentence to be analyzed according to a preset analysis tool with a Chinese recognition function to obtain a sentence pattern structure and at least one target keyword of the sentence to be analyzed, wherein the sentence pattern structure comprises a part-of-speech identifier of each target keyword;

the acquisition module is used for acquiring target data according to the at least one target keyword;

and the display module is used for displaying the target data through a chart set matched with the sentence pattern structure.

11. An electronic device comprising a memory, a processor and a computer program stored on the memory, characterized in that the processor executes the computer program to implement the steps of the method of any of claims 1-9.

12. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 9.

13. A computer program product comprising a computer program, characterized in that the computer program realizes the steps of the method of any one of claims 1-9 when executed by a processor.