Search

KR-102961355-B1 - METHOD FOR GENERATING SUGGESTION VIDEOS TO ASSIST CUSTOMER DECISION MAKING BASED ON ARTIFICIAL INTELLIGENCE

KR102961355B1KR 102961355 B1KR102961355 B1KR 102961355B1KR-102961355-B1

Abstract

The present invention relates to a method for generating suggestion videos to assist artificial intelligence-based customer judgment. The above method is characterized by comprising the steps of: receiving a question from a customer; inferring a question type from the question content; inferring an emotion type from the question content; generating a suggested video for the inferred question type and emotion type; and transmitting the suggested video to the customer.

Inventors

  • 김유석

Assignees

  • 주식회사 일만백만

Dates

Publication Date
20260508
Application Date
20241113

Claims (10)

  1. Step of receiving question content from the customer in the input section; A step in which the question type is inferred from the question content in the inference unit; A step in the inference section to infer the emotion type from the question content; A step of generating a suggestion video for the question type and emotion type inferred in the generation unit; and It includes a step of transmitting a proposed video to the customer from the communications department, and The above inference unit, The content of the customer's questions is classified into multiple question types, including a first category regarding the subject of the question and a second category regarding the purpose of the question, which includes information acquisition, information familiarization, and information analysis. Classify the content of customer questions into multiple emotion types, including happy, sad, cautious, and light emotions, and In the above-mentioned generation unit, multiple video templates are mapped according to question type and emotion type, and each of the multiple video templates has different video lengths classified as long form, middle form, and short form, video brightness, video transition count, and text ratio, The above-mentioned generation unit extracts a video source and a text source according to the first category of the question type, the video source is a video frame constituting the proposed video, and the text source is text that is synthesized into the video frame and displayed together. The above generation unit selects a video template that matches the second category of the question type and the emotion type among a plurality of video templates, and generates a proposed video by encoding the video source and text source to match the video template. The above generating unit is, If the second category of the question type is information analysis, regardless of the sentiment type, generate a suggested video using a video template where the video length is long, the video brightness is medium, the video conversion count is minimum, and the text ratio is maximum, and If the second category of the question type is information comprehension and the emotion type is cautious, a proposed video is generated using a video template with a long-form video length, medium video brightness, medium video conversions, and medium text ratio; if the emotion type is light, a proposed video is generated using a video template with a medium-form video length, medium video brightness, medium video conversions, and minimum text ratio; and if the emotion type is neither cautious nor light, a proposed video is generated using a video template with a medium-form video length, medium video brightness, medium video conversions, and medium text ratio. A method characterized by generating a suggested video using a video template in which the video length is short, the video brightness is maximum, the video conversion count is maximum, and the text ratio is medium when the second category of the question type is information acquisition, and the emotion type is happy, and the video length is short, the video brightness is maximum, the video conversion count is maximum, and the text ratio is medium when the emotion type is sad, and generating a suggested video using a video template in which the video length is short, the video brightness is minimum, the video conversion count is minimum, and the text ratio is minimum when the emotion type is neither happy nor sad.
  2. In paragraph 1, The above inference unit is a method for generating a proposed image using a question type inference model in which the association relationship between multiple question contents and multiple question types is learned, and an emotion type inference model in which the association relationship between multiple question contents and multiple emotion types is learned.
  3. In paragraph 2, The step of inferring the question type from the question content in the above inference unit is, A step of converting the input question content into text and preprocessing it in the preprocessing unit; A step of analyzing morphemes in the preprocessed question content in a morphological analyzer; A step of extracting word frequencies within the question content morphologically analyzed in the word frequency extraction unit; and A method comprising the step of extracting specific keywords from question content based on word frequencies extracted from a keyword extraction unit.
  4. In paragraph 2, A method further comprising the step of receiving feedback on a proposed video from a customer from a customer terminal in the above communication unit.
  5. In paragraph 4, A method further comprising the step of regenerating a suggestion video for the question type and emotion type fed back from the above-mentioned generation unit and sending it to the customer.
  6. In paragraph 4, A method further comprising a step in which the question type inference model and the emotion type inference model are retrained and updated based on the question type and emotion type fed back from the model learning unit.
  7. delete
  8. delete
  9. delete
  10. In paragraph 1, A method in which the question content includes at least one of text, voice data, and video data.

Description

Method for generating suggestion videos to assist customer decision-making based on artificial intelligence The present invention relates to a method for generating a proposed image to assist in the judgment of an AI-based customer. Currently, alongside the advancement of artificial intelligence, technology is being developed to use AI to understand customer inquiries (search requests), generate videos that match the customer's intent, and create suggestion videos to assist in their decision-making. With the advancement of video data, even when conveying the same content, videos can be generated in various formats such as video playback time, information delivery text, and scene transitions. For example, various forms of video data, such as short-form videos with short playback times and long-form videos with long playback times, are attracting people's attention. Accordingly, when generating proposal videos, it is necessary to accurately understand the customer's questions and requests to provide precise proposal videos. Furthermore, if the customer's emotional state is identified and the format of the proposal video is determined and provided based on that emotion, customer needs can be better satisfied. FIG. 1 is a configuration diagram showing a system for performing a method according to one embodiment of the present invention. FIG. 2 is a configuration diagram of an inference unit of a system performing a method according to one embodiment of the present invention. FIG. 3 is a flowchart illustrating an operation method according to one embodiment of the method of the present invention. The present invention is capable of various modifications and may have various embodiments, and some embodiments are illustrated in the drawings and described in detail. However, this is not intended to limit the present invention to specific embodiments, and the scope of the present invention should be understood to include all modifications, variations, equivalents, and substitutions that fall within the technical spirit of the present invention related to the embodiments. Terms such as "first," "second," etc., may be used to describe various components, but said components shall not be limited by said terms. Such terms are used solely for the purpose of distinguishing one component from another. The term "and/or" includes a combination of multiple related described items or any of the multiple related described items. When it is stated that one component is "connected" or "connected" to another component, it should be understood that while it may be directly connected or connected to that other component, there may also be other components in between. It should be understood that no other components exist in between only when it is stated that one component is "directly connected" or "directly connected" to another component. The terms used in this application are used merely to describe the embodiments presented and are not intended to limit the invention. The singular expression includes the plural expression unless the context clearly indicates otherwise. In this application, terms such as "comprising" or "having" are intended to specify the existence of the features, numbers, steps, actions, components, parts, or combinations thereof described in the specification, and should be understood as not precluding the existence or addition of other features, numbers, steps, actions, components, parts, or combinations thereof. As used in this application, "the above" and similar designations may indicate both singular and plural forms. Furthermore, unless there is a description explicitly specifying the order of steps describing the method according to this disclosure, the described steps may be performed in the order intended to achieve the purpose. This disclosure is not limited by the order in which the described steps are described. All terms used in this specification, including technical or scientific terms, have the same meaning as generally understood by those skilled in the art to which the present invention pertains, and should not be interpreted in an ideal or overly narrow formal sense; and where the meaning of any term is defined in this specification, such term shall be interpreted as defined. Hereinafter, preferred embodiments of the present invention will be described in more detail with reference to the attached drawings. In order to facilitate an overall understanding of the present invention, the same reference numerals are used for identical components in the drawings, and redundant descriptions of identical components are omitted. The method (1000) of the present invention is a method for generating a suggestion video to assist a customer in making a judgment, and is characterized by the technical feature of generating a suggestion video by inferring the question type and emotion type regarding the question content received from the customer using artificial intelligence, and setting the length of th