Insurance sales methods and sales devices

TWI935947BActive Publication Date: 2026-08-11CATHAY UNITED BANK
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
TW114131712
Authority / Receiving Office
TW · TW
Patent Type
Patents
Current Assignee / Owner
Filing Date
2025-08-20
Publication Date
2026-08-11
Estimated Expiration
2045-08-19

Smart Images

  • Figure TWG2TB001905886_001
    Figure TWG2TB001905886_001
  • Figure TWG2TB001905886_002
    Figure TWG2TB001905886_002
  • Figure TWG2TB001905886_003
    Figure TWG2TB001905886_003
Patent Text Reader

Abstract

An insurance sales device includes a processing unit, a communication unit electrically connected to the processing unit, and a storage unit to implement an insurance sales method. After the processing unit receives a sales keyword, a risk keyword, and an insurance information image from a user device, it retrieves a sales recording file from the storage unit. Then, it uses a first external website, a second external website, a speech-to-text module, and an image recognition module to obtain an external insurance message, a sales text message, an insurance image message, and a sales risk message. Finally, it obtains multiple scores based on a keyword scoring table and inputs them into a sales testing module to obtain a test probability value.
Need to check novelty before this filing date? Find Prior Art

Claims

1. An insurance sales device, applicable to a user terminal device, a first external website, and a second external website, and comprising: a communication unit for providing connectivity functionality; A storage unit stores a sales history database, a speech-to-text module, an image recognition module, a keyword scoring table, and a sales testing module. The sales history database stores multiple sales records, each of which includes an insurance keyword and a sales recording. The speech-to-text module converts the sales recording into a sales text message. The keyword scoring table includes multiple keyword scores corresponding to multiple scoring keywords. The sales testing module has been trained using multiple sample data and is used to test the probability of sales success to obtain a test probability value. A processing unit is electrically connected to the communication unit and the storage unit to communicate with the user device, the first external website, and the second external website. After receiving a sales keyword, a risk keyword, and an insurance information image from the user device, the processing unit retrieves the sales recording corresponding to the insurance keyword that matches the sales keyword from the sales history database and inputs the sales recording into the speech-to-text module to obtain the sales text message corresponding to the sales recording. The processing unit inputs the insurance information image into the image recognition module to identify the text and scenery in the image and obtain an insurance image message corresponding to the insurance information image; the processing unit uses the communication unit to obtain an external insurance message corresponding to the sales keyword from the first external website using web crawler technology; the processing unit uses the communication unit to obtain a sales risk message corresponding to the risk keyword from the second external website using web crawler technology. The processing unit determines, based on the keyword scoring table, whether the sales text message, the insurance image message, the external insurance message, and the sales risk message contain any of the scoring keywords that match. It then sums up the keyword scores corresponding to the matching keywords to obtain multiple scoring scores for each of the sales text message, the insurance image message, the external insurance message, and the sales risk message. These scoring scores are then input into the sales testing module to obtain the test probability value. When the test probability value is determined to be greater than a preset probability value, the processing unit transmits the sales text message, the insurance image message, the external insurance message, and the sales risk message to the user device via the communication unit.

2. The insurance sales device as described in claim 1, wherein, When the processing unit determines that the test probability value is less than the preset probability value, the processing unit stores the sales text message corresponding to the sales keyword, the external insurance message, the insurance image message, the sales risk message, and the corresponding score and test probability value in the storage unit to add them as sample data.

3. The insurance sales device as claimed in claim 1, wherein, The first external website includes multiple insurance company websites, multiple news websites, multiple hospital websites, and multiple health education websites. The second external website includes a Judicial Yuan website, a Financial Supervisory Commission website, an Insurance Bureau website, an Inspection Bureau website, a Consumer Protection Association website, and a Fair Trade Commission website.

4. The insurance sales device as claimed in claim 1, wherein, Each sales data item also includes at least one sales field information, at least one sales summary tag, and a sales success record. In the processing unit, the communication unit first uses web crawling technology to obtain a first external message corresponding to the sales keyword from the first external website, and uses web crawling technology to obtain a second external message corresponding to the filtered first external message, the at least one sales field information, and the at least one sales summary tag from the first external website. Based on the sales success records, the sales field information of the sales data corresponding to the sales success records is obtained from the sales data. The web crawling technology is then used to obtain the external insurance information of the filtered second external message and the sales field information corresponding to the sales success records from the first external website.

5. The insurance sales device as described in claim 4, wherein, The storage unit also stores a semantic optimization module. After the processing unit obtains the second external information from the first external website using web crawler technology, the processing unit also inputs the filtered second external information into the semantic optimization module to obtain the semantically optimized second external information. The processing unit then uses web crawler technology to obtain the corresponding semantically optimized second external information and the external insurance information of the sales field information corresponding to the sales success records from the first external website.

6. An insurance sales method applicable to an insurance sales device, a user terminal device, a first external website and a second external website, the insurance sales device including a communication unit, a storage unit and a processing unit, the insurance sales method including the following steps: (A) storing a sales history database, a speech-to-text module, an image recognition module, a keyword scoring table and a sales test module in the storage unit, the sales history database storing multiple sales records, each of the sales records including an insurance keyword and a sales recording file, the speech-to-text module being used to convert the sales recording file into a sales text message, the keyword scoring table including multiple keyword scores corresponding to multiple scoring keywords, and the sales test module having been trained with multiple sample data and used to test the probability of sales success to obtain a test probability value; (B) The processing unit communicates with the user device, the first external website, and the second external website via the communication unit. After receiving a sales keyword, a risk keyword, and an insurance information image from the user device, it retrieves the sales recording file corresponding to the insurance keyword that matches the sales keyword from the sales history database and inputs the sales recording file into the speech-to-text module to obtain the sales text message corresponding to the sales recording file; (C) The processing unit inputs the insurance information image into the image recognition module to recognize the text and scenery in the image to obtain an insurance image message corresponding to the insurance information image; (D) The processing unit obtains an external insurance message corresponding to the sales keyword from the first external website via the communication unit using web crawler technology. (E) The processing unit obtains a sales risk message corresponding to the risk keyword from the second external website using web crawler technology via the communication unit; and (F) The processing unit determines, according to the keyword scoring table, whether the sales text message, the insurance image message, the external insurance message, and the sales risk message contain any of the scoring keywords, and sums up the keyword scores corresponding to the matching keywords to obtain multiple scoring scores corresponding to the sales text message, the insurance image message, the external insurance message, and the sales risk message, respectively. The scoring scores are then input into the sales test module to obtain the test probability value. When the test probability value is determined to be greater than a preset probability value, the sales text message, the insurance image message, the external insurance message, and the sales risk message are transmitted to the user device via the communication unit.

7. The insurance sales method as described in claim 6, wherein, The first external website includes multiple insurance company websites, multiple news websites, multiple hospital websites, and multiple health education websites. The second external website includes a Judicial Yuan website, a Financial Supervisory Commission website, an Insurance Bureau website, an Inspection Bureau website, a Consumer Protection Association website, and a Fair Trade Commission website.

8. The insurance sales method as described in claim 6, wherein, In step (F), when the processing unit determines that the test probability value is less than the preset probability value, the scoring scores corresponding to the sales text message, the external insurance message, the insurance image message, and the sales risk message corresponding to the sales keyword, as well as the test probability value, are stored in the storage unit to be added as the sample data.

9. The insurance sales method as described in claim 6, wherein, In step (A), each sales data item further includes at least one sales field information, at least one sales summary tag, and a sales success record. In step (D), the processing unit first obtains a first external message corresponding to the sales keyword from the first external website via the communication unit using web crawler technology, and obtains a second external message corresponding to the filtered first external message, the at least one sales field information, and the at least one sales summary tag from the first external website using web crawler technology. Based on the sales success records, the processing unit obtains the sales field information of the sales data corresponding to the sales success records from the sales data, and obtains the external insurance information of the filtered second external message and the sales field information corresponding to the sales success records from the first external website using web crawler technology.

10. The insurance sales method as described in claim 9, wherein, In step (A), the storage unit also stores a semantic optimization module. In step (D), after the processing unit obtains the second external information from the first external website using web crawler technology, it also inputs the filtered second external information into the semantic optimization module to obtain the semantically optimized second external information. Then, it uses web crawler technology to obtain the corresponding semantically optimized second external information and the external insurance information of the sales field information corresponding to the sales success records from the first external website.

Citation Information

Patent Citations

  • Insurance service recommendation method and device, equipment and storage medium

    CN118193843A

  • Insurance selling device

    TWM680869U

  • Systems and methods for collecting and processing alternative data sources for risk analysis and insurance

    US20220005125A1