High-correlation intelligent service recommendation method

Abstract

The invention discloses a high-correlation intelligent service recommendation method, which comprises the following steps of: extracting data of documents historically filled by a user, and classifying the data; extracting Chinese text information in the data, and obtaining word vectors of the Chinese text information; carrying out normalization processing on the word vectors of the Chinese text information and other information; performing clustering and impurity removal on the normalized data; and service recommendation is carried out when the user fills in the document. According to the high-correlation intelligent service recommendation method with the above steps, the core Chinese fields in the historical form filling information of the user are processed through the artificial neural network to obtain dense word vectors, and then clustering impurity removal is utilized to remove noisy point data, so that the recommended service is more in line with a result wanted by the user, the form filling difficulty is reduced. Error information input is reduced, and filling efficiency is improved.

Description

technical field [0001] The invention belongs to the fields of data analysis and data recommendation, and in particular relates to a service recommendation method. Background technique [0002] With the rapid development and widespread popularization of mobile Internet in the world, more and more traditional industries have been challenged unprecedentedly. importance in construction. [0003] Based on traditional data analysis, it is difficult to meet the needs of the current environment. Traditional data analysis mainly includes statistics, big data analysis, etc.; business recommendation technology solutions based on traditional statistics can only handle statistics when each field is the same , when recommending solutions for smart services, there are often many recommended field types, and when it comes to Chinese text statistics, it cannot be divided into statistics in the same dimension according to word segmentation and word vector conversion. Business recommendation...

AI Technical Summary

Problems solved by technology

[0003] Based on traditional data analysis, it is difficult to meet the needs of the current environment. Traditional data analysis mainly includes statistics, big data analysis, etc.; business recommendation technology solutions based on traditional statistics can only handle statistics when each field is the same , when recommending solutions for smart services, there are often many recommended field types, and when it comes to Chinese text statistics, it cannot be divided into statistics in the same dimension according to word segmentation and word vector conversion.

Business recommendations based on traditional statistics are often not accurate and flexible enough, and are quite different from the results desired by users, making it impossible to recommend ideal documents for users

Images