A deep learning-based multi-modal data quality evaluation method

By employing a deep learning-based multimodal data quality assessment method, which utilizes semantic feature extraction and cross-modal joint encoding, the limitations of traditional methods in multimodal data assessment are overcome, achieving highly accurate data quality assessment and providing fine-grained assessment capabilities.

CN120804084BActive Publication Date: 2025-11-21GUANGZHOU PRINCIPAL DATA CO LTD
2 Cites 0 Cited by

Patent Information

Application Number
CN202511299320.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2025-09-12
Publication Date
2025-11-21
Estimated Expiration
2045-09-12

AI Technical Summary

Technical Problem

Traditional database data quality assessment methods have limitations when dealing with multimodal data. They are difficult to adapt to the diversity and dynamic changes of multimodal data, and lack fine-grained hash cross-modal coding and bidirectional reconstruction offset assessment, resulting in a lack of accuracy in the assessment methods.

Method used

A deep learning-based approach is adopted to extract semantic features and perform cross-modal joint encoding by acquiring first-modal and second-modal record data labeled as aligned in the database. Feature extraction is performed using MobileNet and LSTM networks, and a structured-metadata cross-modal quality assessment joint encoded feature tensor is generated by combining a fine-grained hashing algorithm and a cross-modal mask weaving network. Data consistency offset is calculated and quality assessment is performed using a deep learning model.

Benefits of technology

It improves the accuracy of multimodal data quality assessment, realizes an end-to-end deep learning framework, solves the problem of the separation between structured data and text description in traditional methods, and provides fine-grained data quality assessment capabilities.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The application discloses a kind of based on deep learning multi-modal data quality evaluation method, comprising obtaining database in first modality record data and second modality record data labeled as alignment;First modality record data and second modality record data are respectively carried out semantic feature extraction to obtain text structured data semantic encoding feature tensor and metadata description semantic encoding feature vector;Text structured data semantic encoding feature tensor and metadata description semantic encoding feature vector cross-modal joint coding to obtain structured-metadata cross-modal quality evaluation joint coding feature tensor;Based on structured-metadata cross-modal quality evaluation joint coding feature tensor, reconstruct record data is generated, and the offset feature of reconstruct record data is calculated;Offset feature is evaluated by quality by first deep learning model;The application improves the accuracy of data quality evaluation by semantic encoding feature extraction, fine-grained hash cross-modal coding and bidirectional reconstruction offset evaluation.
Need to check novelty before this filing date? Find Prior Art