Word weight prediction model generation method, job recommendation method and computing equipment

Abstract

The invention discloses a word weight prediction model generation method, a job recommendation method and a computing device for weight prediction of words in job titles. The word weight prediction model generation method includes: obtaining a pre-collected job title training set, job title The title training set includes multiple job titles, and each job title is associated with a corresponding word segmentation sequence. The word segmentation sequence includes one or more words obtained after word segmentation of the job title; for each job title, obtain its associated word segmentation sequence The word features of each word; calculate the marking weight of each word in the word segmentation sequence associated with each job title; construct a logistic regression model; take the word feature of each word in the word segmentation sequence associated with each job title as input, and use the corresponding marking weight As output, the logistic regression model is trained to generate a term weight prediction model whose output indicates the predicted weight of each term in the input job title.

Description

technical field [0001] The present invention relates to the field of Internet technology, in particular to a word weight prediction model generation method, a job recommendation method and a computing device for weight prediction of words in job titles. Background technique [0002] With the rapid development of Internet technology, more and more people are beginning to enjoy various conveniences that the Internet brings to work and life. Taking the traditional recruitment industry as an example, people can now use recruitment websites to search for companies and positions they are interested in, and then choose whether to submit resumes or contact employers. [0003] After the user enters the search term on the search page of the recruitment website, the search navigation can help the user quickly find the required job information among the massive job positions, and recommend each job information to the user in a certain order. At present, the mainstream job recommendatio...

AI Technical Summary

Problems solved by technology

At present, the mainstream job recommendation method mostly calculates the correlation between the search term and the job information, and feeds back the corresponding job information to the user according to the order of the correlation from high to low, and users generally give priority to browsing and ranking first. It is easy to ignore the following job information, which requires high accuracy and short time-consuming correlation calculation. However, the existing job recommendation methods have a big deficiency in this aspect.

Images