Collaborative filtering-based environmental pollution loss estimation method

Abstract

The invention discloses a collaborative filtering-based environmental pollution loss estimation method. The method comprises the steps of conducting the normalization treatment on the emission amount of the loss factor industrial wastewater, the emission amount of the industrial chemical oxygen demand, the industrial nitrogen oxygen emission amount, the urban domestic sewage emission mount, the daily chemical oxygen demand amount and the daily nitrogen oxygen emission amount; calculating the similarity degree: calculating the similarity degree among sewage discharge scenarios in all cities based on a Pearson correlation coefficient; selecting neighboring cities Top K: based on the similarity degree obtained through calculation, searching k cities similar in sewage discharge scenario to a target city, wherein the similarity degree among all the neighboring cities is larger than 0; estimating the economic loss of the target city based on the similarity degree obtained through calculation and the Top K neighboring cities. The method is simple and convenient to operate, wherein the problems of existing collaborative filtering recommendation methods can be well solved.

Description

technical field [0001] The invention belongs to the technical field of environmental pollution loss estimation, in particular to a method for estimating environmental pollution loss based on collaborative filtering. Background technique [0002] The recommendation system first appeared in the 1990s. With the proposal and development of the collaborative filtering method, the recommendation system has received extensive attention in the fields of e-commerce, social finance, and network psychology, and has become a popular research direction. In recent years, recommender systems have been maturely used in computer science, management science and engineering, data mining, e-commerce and other fields. The methods mainly include: 1) Collaborative Filtering Recommendation method (Collaborative Filtering Recommendation): According to the comparison of item preferences between other users and target users, similar users are selected. Then, through the analysis of the item selection...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a method for estimating environmental pollution losses based on collaborative filtering, which aims to solve the problem that existing collaborative filtering recommendation methods are difficult to evaluate accurate economic loss values, and the loss weights obtained by training on different data may be different. question

Images