Search

CN-121998775-A - Data prediction method and device based on artificial intelligence, computer equipment and medium

CN121998775ACN 121998775 ACN121998775 ACN 121998775ACN-121998775-A

Abstract

The application belongs to the technical field of artificial intelligence, and relates to a data prediction method based on artificial intelligence, which comprises the steps of receiving typhoon name information input by a user; the method comprises the steps of collecting relevant text information of target typhoons corresponding to typhoon name information, abstracting the relevant text information based on a target big model to obtain abstract text, encoding the collected numerical data related to the target typhoons to obtain target numerical data, integrating the abstract text and the target numerical data to obtain multi-modal information, analyzing the multi-modal information based on a typhoon loss prediction model to obtain a prediction result, and outputting the prediction result. The application also provides a data prediction device, computer equipment and a storage medium based on the artificial intelligence. In addition, the prediction results of the present application may be stored in the blockchain. The typhoon loss prediction method and the typhoon loss prediction device can be applied to typhoon loss prediction scenes in the field of financial science and technology, the processing accuracy of typhoon loss prediction is improved, and the accuracy of the generated prediction result is ensured.

Inventors

  • Kong Lingge

Assignees

  • 中国平安财产保险股份有限公司

Dates

Publication Date
20260508
Application Date
20260113

Claims (10)

  1. 1. An artificial intelligence-based data prediction method is characterized by comprising the following steps: Receiving typhoon name information input by a user; collecting relevant text information of a target typhoon corresponding to the typhoon name information, and performing abstract generation processing on the relevant text information based on a preset target big model to obtain a corresponding abstract text; Collecting numerical data relating to the target typhoons; performing coding processing on the numerical data based on a preset coding strategy to obtain corresponding target numerical data; Integrating the abstract text and the target numerical data to obtain corresponding multi-mode information; Analyzing and processing the multi-mode information based on a preset typhoon loss prediction model to obtain a corresponding prediction result; And outputting the prediction result.
  2. 2. The method for predicting data based on artificial intelligence according to claim 1, wherein the step of collecting relevant text information of a target typhoon corresponding to the typhoon name information, and performing abstract generation processing on the relevant text information based on a preset target big model to obtain corresponding abstract text comprises the following steps: based on a preset inquiry expansion strategy, carrying out information crawling on the typhoon name information to obtain corresponding related text information; cleaning the related text information based on a preset cleaning strategy to obtain corresponding target text information; based on a preset prompt template, performing abstract generation processing on the target text information by using the target large model to obtain corresponding abstract information; optimizing the abstract information based on a preset optimizing strategy to obtain corresponding target abstract information; And taking the target abstract information as the abstract text.
  3. 3. The method for predicting data based on artificial intelligence according to claim 2, wherein the step of crawling information of typhoon name information based on a preset query expansion strategy to obtain corresponding related text information specifically comprises: inputting the typhoon name information to a preset web crawler to acquire query data related to the target typhoon from a network; generating a specified number of query terms based on the query data; Performing network crawling processing on a preset data source based on the query word to obtain corresponding crawling information; and taking the crawling information as the text information.
  4. 4. The artificial intelligence-based data prediction method according to claim 2, wherein the step of optimizing the summary information based on a preset optimization strategy to obtain the corresponding target summary information specifically comprises: performing accuracy optimization processing on the abstract information to obtain first abstract information; carrying out integrity optimization processing on the first abstract information to obtain corresponding second abstract information; Carrying out succinct optimization processing on the second abstract information to obtain corresponding third abstract information; Performing logic optimization processing on the third abstract information to obtain corresponding fourth abstract information; Performing readability optimization processing on the fourth abstract information to obtain corresponding fifth abstract information; and taking the fifth abstract information as the target abstract information.
  5. 5. The artificial intelligence-based data prediction method according to claim 1, wherein the step of performing encoding processing on the numerical data based on a preset encoding policy to obtain corresponding target numerical data specifically comprises: normalizing the numerical data to obtain corresponding first processing data; Discretizing the first processing data to obtain corresponding second processing data; carrying out serialization processing on the second processing data to obtain corresponding third processing data; And taking the third processing data as the target numerical value data.
  6. 6. The artificial intelligence based data prediction method according to claim 1, wherein the step of integrating the abstract text and the target numerical data to obtain the corresponding multi-modal information specifically comprises: converting the abstract text based on a preset text word segmentation device to obtain a corresponding text sequence; Acquiring a preset splicing strategy; performing splicing processing on the text sequence and the target numerical data based on the splicing strategy to obtain corresponding spliced data; And taking the spliced data as the multi-mode information.
  7. 7. The artificial intelligence-based data prediction method according to claim 1, further comprising, before the step of analyzing the multi-modal information based on the preset typhoon loss prediction model to obtain a corresponding prediction result: acquiring historical typhoon data collected in advance; Carrying out sample construction processing on the historical typhoon data to obtain corresponding sample data; Calling a preset base large model; training and optimizing the base large model by using the sample data based on a preset model training strategy until a specified model meeting the construction requirement is obtained; and taking the appointed model as the typhoon loss prediction model, and carrying out deployment processing on the typhoon loss prediction model.
  8. 8. An artificial intelligence based data prediction apparatus, comprising: The receiving module is used for receiving typhoon name information input by a user; the generation module is used for collecting relevant text information of the target typhoon corresponding to the typhoon name information, and carrying out abstract generation processing on the relevant text information based on a preset target big model to obtain a corresponding abstract text; The collection module is used for collecting numerical data related to the target typhoons; the encoding module is used for encoding the numerical data based on a preset encoding strategy to obtain corresponding target numerical data; The integration module is used for integrating the abstract text and the target numerical data to obtain corresponding multi-mode information; The analysis module is used for analyzing and processing the multi-mode information based on a preset typhoon loss prediction model to obtain a corresponding prediction result; And the output module is used for outputting and processing the prediction result.
  9. 9. A computer device comprising a memory having stored therein computer readable instructions which when executed implement the steps of the artificial intelligence based data prediction method of any of claims 1 to 7.
  10. 10. A computer readable storage medium having stored thereon computer readable instructions which when executed by a processor implement the steps of the artificial intelligence based data prediction method of any of claims 1 to 7.

Description

Data prediction method and device based on artificial intelligence, computer equipment and medium Technical Field The application relates to the technical field of artificial intelligence, which can be applied to the fields of financial science and technology, digital medical treatment and the like, in particular to a data prediction method, a data prediction device, computer equipment and a storage medium based on artificial intelligence. Background In the field of insurance, the loss caused by natural disasters, especially typhoons, is extremely large in scale, and brings significant pressure to the reimbursement of insurance companies. The loss caused by the accurate prediction of typhoon and the risk of the payoff are effectively reduced to be the key problems to be solved urgently in the insurance industry, wherein the accurate prediction of typhoon loss is a key link. At present, the main stream technical route aiming at typhoon loss prediction in the protection field mainly adopts a traditional machine learning and deep learning-based method, and the method is the most widely applied technical route in the current industry. The technical path is typically characterized by historical typhoons' meteorological data (e.g., maximum wind speed, central air pressure, speed of movement, path, etc.) and geographic socioeconomic data (e.g., population density, GDP, infrastructure distribution, etc.) as inputs, and the loss amount or risk level is predicted by training regression or classification models. However, such methods have many drawbacks and disadvantages, including a single information modality and insufficient data utilization. Such models are essentially "numerical models" whose inputs are strictly defined within a structured numerical table. In practical applications, a large amount of very valuable unstructured information, in particular text information from internet news, social media, government reports and expert analysis, is completely excluded from the model. The model can only stay on the surface of the model to realize the cognition of typhoon influence, and a comprehensive and deep typhoon image cannot be constructed, so that the comprehensive influence of typhoons in all aspects is difficult to accurately grasp. Secondly, the static analysis capability is insufficient, and trend insight is lacking. Traditional models learn and predict based on static historical data, essentially a "static correlation" rather than a "dynamic deduction. In the typhoon development process, risk factors are in a dynamic evolution state, and the whole process comprises rich trend information from early warning release to disaster login to post-disaster influence. However, the conventional model is difficult to understand and model the dynamic evolution rules, and the trend information in the complete chain cannot be effectively captured. Therefore, the method has weak capability in trend prejudgment and prospective analysis, so that typhoon loss prediction accuracy is low, and the requirement of the insurance industry on accurate prediction cannot be met. In view of this, there is a need to develop a new technique to overcome the deficiencies of the prior art and improve the accuracy and reliability of typhoon loss prediction. Disclosure of Invention The embodiment of the application aims to provide a data prediction method, a device, computer equipment and a storage medium based on artificial intelligence, so as to solve the technical problem of lower accuracy in the traditional typhoon loss prediction method based on traditional machine learning and deep learning. In a first aspect, there is provided an artificial intelligence based data prediction method, comprising: Receiving typhoon name information input by a user; collecting relevant text information of a target typhoon corresponding to the typhoon name information, and performing abstract generation processing on the relevant text information based on a preset target big model to obtain a corresponding abstract text; Collecting numerical data relating to the target typhoons; performing coding processing on the numerical data based on a preset coding strategy to obtain corresponding target numerical data; Integrating the abstract text and the target numerical data to obtain corresponding multi-mode information; Analyzing and processing the multi-mode information based on a preset typhoon loss prediction model to obtain a corresponding prediction result; And outputting the prediction result. In a second aspect, there is provided an artificial intelligence based data prediction apparatus comprising: The receiving module is used for receiving typhoon name information input by a user; the generation module is used for collecting relevant text information of the target typhoon corresponding to the typhoon name information, and carrying out abstract generation processing on the relevant text information based on a preset target big model to obtain a correspo