Search

CN-122021643-A - Interview text analysis method and device and electronic equipment

CN122021643ACN 122021643 ACN122021643 ACN 122021643ACN-122021643-A

Abstract

The invention discloses an interview text analysis method and device and electronic equipment, and relates to the field of financial science and technology or other related fields, wherein the method comprises the steps of determining at least one key character from a target text set to be analyzed; the method comprises the steps of analyzing at least one key character through a text analysis model to obtain an analysis result matched with a target text set, determining a user portrait and a user trip chart matched with the target text set based on the analysis result, and determining a demand analysis result matched with the target text set based on the user portrait and the user trip chart. The invention solves the technical problem of lower analysis efficiency of interview texts in the related technology.

Inventors

  • Hui Chuhang

Assignees

  • 中国工商银行股份有限公司

Dates

Publication Date
20260512
Application Date
20260119

Claims (10)

  1. 1. A method of analyzing interview text, comprising: determining at least one key character from a target text set to be analyzed, wherein the target text set comprises an interview text set, and at least one key character is matched with the target text set; Analyzing at least one key character through a text analysis model to obtain an analysis result matched with the target text set, wherein the text analysis model is used for analyzing the relation between at least one reference character; Determining a user portrait and a user trip chart matched with the target text set based on the analysis result, wherein the user portrait is used for indicating the user requirement matched with the target text set, the user trip chart is used for indicating the behavior path and emotion change of a user when using a target product, and the analysis result is used for indicating the clustering grouping result of at least one key character; And determining a demand analysis result matched with the target text set based on the user portrait and the user trip chart.
  2. 2. The method of claim 1, further comprising, prior to said analyzing at least one of said key characters by a text analysis model to obtain an analysis result matching said target text set: Determining at least one reference character matched with the reference text set from the reference text set; And acquiring a semantic vector corresponding to each reference character, and acquiring the text analysis model based on at least one semantic vector.
  3. 3. The method of claim 2, wherein the obtaining the text analysis model based on at least one of the semantic vectors comprises: Determining at least one reference character as a reference node on a correlation graph, wherein a target character in the at least one reference character corresponds to the target reference node on the correlation graph; Under the condition that at least one reference character comprises a first character and a second character, obtaining a similarity coefficient between a first semantic vector corresponding to the first character and a second semantic vector of the second semantic character, wherein the similarity coefficient is used for indicating semantic similarity between the first character and the second character; Under the condition that the similarity coefficient meets a similarity condition, connecting a first node corresponding to a first character and a second node corresponding to a second character on the association graph through a reference connection edge, and configuring reference weight for the reference connection edge, wherein the reference weight is matched with the similarity coefficient; And inputting the association diagram into an initial text analysis model for training until the trained text analysis model is obtained.
  4. 4. The method of claim 1, wherein the determining at least one key character from the set of target text to be analyzed comprises: inputting the target text set into a semantic segmentation model to obtain at least one semantic unit; under the condition that a target semantic unit in at least one semantic unit is matched with a target character in a preset key character set, determining the target character as the key character; And under the condition that a moving object relation exists between a first semantic unit and a second semantic unit in at least one semantic unit, determining the first semantic unit, the second semantic unit and the moving object relation as the key character, wherein the moving object relation is used for indicating grammar association formed by verbs and objects.
  5. 5. The method of claim 1, wherein determining a user representation and a user itinerary that match the target text set based on the analysis results comprises: determining at least one user identity tag matched with the target text set, and inputting the target text set into an emotion analysis model to obtain at least one emotion tag matched with the target text set; And determining the user portrait and the user trip chart matched with the target text set through at least one user identity label, at least one emotion label and the analysis result.
  6. 6. The method of claim 5, wherein said determining said user representation and said user itinerary matching said target text set from at least one of said user identity tag and at least one of said emotion tag, and said analysis result, comprises: Determining at least one group of character sets matched with a reference identity tag from the analysis result, wherein the character sets are obtained by clustering and grouping based on semantic similarity among at least one key character, and the at least one user identity tag comprises the reference identity tag; Determining at least one target emotion label matched with the reference identity label from at least one emotion label; Carrying out weighted fusion on the reference identity tag, at least one target emotion tag and at least one group of character set to obtain a user image matched with the reference identity tag; And under the condition that at least one behavior character describing the behavior of the user exists in at least one group of character sets, connecting behavior nodes corresponding to the at least one behavior character according to a behavior flow relation, and marking at least one target emotion label to the behavior nodes to obtain the user trip map.
  7. 7. The method of any of claims 1 to 6, wherein the determining a demand analysis result that matches the target text set based on the user representation and the user trip map comprises at least one of: acquiring a first analysis result when the user portrait meets a first portrait condition and the user trip chart meets a first behavioral condition, wherein the first analysis result is used for indicating that the guidance and operation instruction needs to be enhanced; And under the condition that the user portrait meets a second portrait condition and the user trip chart meets a second behavior condition, acquiring a second analysis result, wherein the second analysis result is used for indicating that the functional compatibility needs to be improved.
  8. 8. An apparatus for analyzing interview text, comprising: A first determining unit, configured to determine at least one key character from a target text set to be analyzed, where the target text set includes an interview text set, and at least one key character matches the target text set; the analysis unit is used for analyzing at least one key character through the text analysis model to obtain an analysis result matched with the target text set, wherein the text analysis model is used for analyzing the relation between at least one reference character; the second determining unit is used for determining a user portrait and a user trip chart matched with the target text set based on the analysis result, wherein the user portrait is used for indicating the user requirement matched with the target text set, the user trip chart is used for indicating the behavior path and emotion change of a user when using a target product, and the analysis result is used for indicating the clustering grouping result of at least one key character; And the third determining unit is used for determining a requirement analysis result matched with the target text set based on the user portrait and the user trip chart.
  9. 9. An electronic device comprising one or more processors and memory for storing one or more programs, wherein the one or more programs, when executed by the one or more processors, cause the one or more processors to implement the method of analysis of interview text of any of claims 1-7.
  10. 10. A computer program product comprising a computer program, characterized in that the computer program, when executed by a processor, implements the steps of the method of analysing interview text according to any one of claims 1 to 7.

Description

Interview text analysis method and device and electronic equipment Technical Field The invention relates to the field of financial science and technology or other related fields, in particular to an interview text analysis method and device and electronic equipment. Background In the field of user experience design and product research, user interviews as a core quality research method to obtain the user's real needs and feedback can produce large amounts of unstructured interview text data. Traditional analysis methods rely primarily on researchers to manually transcribe and read text, and the overall process is highly dependent on personal experience and time consuming and laborious, resulting in inefficient analysis of interview text. It will be appreciated that the prior art has a technical problem of less efficient analysis of interview text. In view of the above problems, no effective solution has been proposed at present. Disclosure of Invention The embodiment of the invention provides an interview text analysis method and device and electronic equipment, which are used for at least solving the technical problem of low analysis efficiency of interview texts in the related technology. In order to achieve the above object, according to one aspect of the present application, there is provided an interview text analysis method, comprising determining at least one key character from a target text set to be analyzed, wherein the target text set contains interview text sets, at least one key character is matched with the target text set, analyzing the at least one key character through the text analysis model to obtain an analysis result matched with the target text set, wherein the text analysis model is used for analyzing a relation between at least one reference character, determining a user portrait and a user trip map matched with the target text set based on the analysis result, wherein the user portrait is used for indicating a user requirement matched with the target text set, the user trip map is used for indicating a behavior path and an emotion change of a user when using a target product, and the analysis result is used for indicating a clustering result of at least one key character, and determining a requirement analysis result matched with the target text set based on the user portrait and the user trip map. According to another aspect of the embodiment of the present invention, there is provided an interview text analysis apparatus, including a first determining unit configured to determine at least one key character from a target text set to be analyzed, wherein the target text set includes an interview text set, at least one key character matches the target text set, an analyzing unit configured to analyze the at least one key character through the text analysis model to obtain an analysis result matching the target text set, wherein the text analysis model is configured to analyze a relationship between at least one reference character, and a second determining unit configured to determine a user portrait and a user trip map matching the target text set based on the analysis result, wherein the user portrait is configured to indicate a user demand matching the target text set, the user trip map is configured to indicate a behavior path and an emotion change of a user when using a target product, and the analysis result is configured to indicate a clustering result of at least one key character, and a third determining unit configured to determine a user portrait and a user trip map matching the target text set based on the analysis result. According to another aspect of an embodiment of the present invention, there is also provided a computer readable storage medium including a stored computer program, wherein the computer program, when run, controls a device on which the computer readable storage medium resides to perform the method of analyzing interview text of any one of the above. According to another aspect of embodiments of the present invention, there is also provided an electronic device comprising one or more processors and a memory for storing one or more programs, wherein the one or more programs, when executed by the one or more processors, cause the one or more processors to implement the method of analyzing interview text of any of the above. According to another aspect of embodiments of the present invention, there is also provided a computer program product comprising a computer program which, when executed by a processor, implements the steps of the method of analysing interview text of any one of the above. According to the embodiment provided by the application, the key characters are automatically extracted from the interview text set, the semantic relation among the characters is analyzed and clustered by utilizing the text analysis model, further, the user portraits representing the user demands and the user trip charts reflecting the behavior paths and the emotion