A multi-person collaborative review method based on an offline networking environment comprises the steps that firstly, performing data importing on external multi-batch data sources, completing dataintegration
processing by analyzing data structures and associations of data sets, and fusing the multiple data sets into a complete structured
data set; then, according to personalized attributes ofdifferent review experts, automatically analyzing content features of the to-be-reviewed
data set, and performing association evaluation and automatic matching with expert attribute features to realize automatic division of review contents; in the evaluation process, after an evaluation expert audits and adjusts the data, automatically completing
data integration and aggregation, data associationcalculation and hierarchical summary calculation; and finally, carrying out word segmentation and semantic analysis on review opinions filled by experts, extracting keywords and quantized data in thetext content, carrying out data summarization calculation and content integration
processing on the multi-person review content on the basis of the identifiers of the content attributes, and automatically generating a review result report.