Intelligent election push broadcast machine model equipment and system
Patent Information
- Application Number
- TW115204786
- Authority / Receiving Office
- TW · TW
- Patent Type
- Utility models
- Current Assignee / Owner
- Filing Date
- 2026-05-26
- Publication Date
- 2026-10-01
- Estimated Expiration
- 2036-05-25
Smart Images

Figure TWG2TB001914393_001 
Figure TWG2TB001914393_002 
Figure TWG2TB001914393_003
Abstract
Claims
1. A smart election push notification machine model device and system, characterized in that it comprises: an AI intelligent management device, including a political opinion data import module and a war room dashboard module, for inputting election political opinion data in text form and monitoring data in real time; an AIGC political opinion generation and processing device, communicatively connected to the AI intelligent management device, including a text semantic parsing module, a script automatic arrangement module, a multimedia generation engine, a virtual avatar model, and an audio-visual synthesis and packaging module, used to receive the election political opinion data and automatically generate virtual human images with lip-sync and customized audio-visual files in multiple languages or styles from a single text script; and an AI algorithm-based precise push notification device, communicatively connected to the AIGC political opinion generation and processing device. The system comprises an audience profiling module and a collaborative filtering push algorithm module, accessing a large database of user data containing the digital profile features and preference tags of all target voters, and using natural language processing algorithms to filter and push corresponding customized video and audio files; and an AI interactive receiving and feedback device, held by voters and communicating with the AI algorithm's precise push device, including a video and audio receiving and decoding module, a user interface interaction module, an AI dialogue assistant module, and a data feedback communication module, receiving and playing the customized video and audio files, and the AI dialogue assistant module allows voters to ask questions and generates real-time question-and-answer responses using search-enhanced generation technology; this system integrates the above four devices to form a complete AI-Driven system. The Election Lifecycle system operates according to a standardized process of "collection-processing-distribution-feedback". The strategic goals are defined by the AI intelligent management device, which drives the AIGC policy generation and processing device to generate diverse digital content. The content is then accurately pushed to voters' devices through AI algorithms, and finally the voters' interactive behavior data is sent back to the AI interactive receiving and feedback device to complete the system's dynamic self-calibration and optimization.
2. The smart election push machine model device and system as described in claim 1, wherein the AIGC policy generation and processing device further integrates a multilingual translation and cultural conversion model, and can automatically convert election policy data in Traditional Chinese form into customized audio and video files containing Taiwanese, Hakka, new immigrant speech, or specific idioms and accents of mainland spouses according to preset language parameters.
3. The smart election push notification machine model device and system as described in claim 1, wherein the AI dialogue assistant module in the AI interactive receiving and feedback device further includes a voice / text recognition unit, a RAG retrieval enhancement unit and a sentiment analysis unit, thereby analyzing the positive and negative emotions of voters when asking questions and transmitting the question records and sentiment analysis data back to the big data user database in real time.
4. The smart election push notification machine model device and system as described in claim 1, wherein the war room dashboard module in the AI intelligent management device can read the real-time interaction and sentiment data in the big data user database to provide candidates with a basis for dynamically adjusting their campaign strategies and pushing notifications a second time.
