Website text content-based online loan website entity recognition method and system

Abstract

The invention provides a website text content-based online loan website entity recognition method and system. The method comprises the steps of S01, constructing a training set domain name table; s02,constructing a prediction set domain name table; s03, data cleaning and preprocessing; s04, training a text classification model to obtain a target text classification model; s05, online loan websiterecognition: inputting the target webpage content field of each sample in the prediction set domain name table into a value target text classification model, and outputting whether the field corresponding to each sample is an online loan website field or not; s06, training a named entity recognition model to obtain a target named entity recognition model; and S07, entity name labeling. The methodis based on operator DPI data. The website domain name host accessed by the user is obtained, the webpage content is obtained, the online loan website is identified. Meanwhile, the entity name in theonline loan website is extracted by utilizing the named entity recognition technology, and then some bad websites are marked and an enterprise blacklist library is established in combination with external blacklist data, so that the method is high in accuracy and high in timeliness.

Description

technical field [0001] The invention relates to the technical field of online loan website identification, in particular to a method and system for identifying an online loan website entity based on website text content. Background technique [0002] With the rapid development of Internet technology, the establishment of websites is more convenient and the threshold is lower, which has led to the emergence of many bad and illegal websites such as illegal online loan websites, phishing websites, and gambling websites. In recent years, incidents such as P2P companies running away, network fraud, and telecom fraud have occurred frequently, causing relatively serious property losses to netizens, some even endangering personal safety, and at the same time producing adverse social impacts. Accurate and efficient identification of online loan websites, and timely remind users to operate carefully, can avoid the loss of users' property, etc., and at the same time improve the company...

AI Technical Summary

Problems solved by technology

[0005] The manual labeling method is costly and inefficient. When the sample size is small, although the classification can be completed accurately, when faced with a large amount of text, it is difficult to identify whether the website is an online loan website only by manual labeling

Recognition methods such as simple dictionaries, rule matching, and machine learning models have low accuracy and poor timeliness, and need to rely on more professional and accurate thesaurus

Images