A method and system for identifying foodborne disease outbreaks based on link prediction

Abstract

The invention relates to a method and system for identifying foodborne disease outbreaks based on link prediction. The method includes: performing data processing on the foodborne disease outbreak data; sampling the case data after the data processing to obtain a set of positive and negative sample pairs; performing feature extraction on each positive and negative sample pair in the positive and negative sample pair set; The positive and negative sample pairs after feature extraction are input into the link prediction model based on the neural network to learn the relationship between cases; according to the output of the link prediction model, a case relationship network is constructed. The nodes in the network represent case entities, and the edges Represents the relationship between cases, and the edge weight reflects the strength of the relationship between cases; according to the case relationship network, the community discovery algorithm is used to obtain foodborne disease outbreaks. The invention transforms the traditional clustering problem into the problem of predicting the correlation between cases and finding the community in the graph network, thereby achieving a better outbreak identification effect than the traditional clustering algorithm.

Description

technical field [0001] The invention belongs to the field of information technology, and relates to a foodborne disease outbreak identification technology, in particular to a link prediction-based foodborne disease outbreak identification method and system. Background technique [0002] Foodborne diseases refer to infectious and toxic diseases caused by pathogenic factors in food entering the human body, including food poisoning, suspected foodborne abnormal diseases and foodborne infectious diseases. Foodborne illnesses threaten human health and cause economic losses worldwide every year. In 2015, the World Health Organization stated that foodborne diseases pose a significant global burden. About 600 million cases of foodborne illness occur worldwide each year, resulting in 420,000 deaths. Therefore, it is necessary to study the surveillance and prevention of foodborne diseases. The identification of foodborne disease outbreaks is an important part of foodborne disease m...

AI Technical Summary

Problems solved by technology

[0009] At present, my country's foodborne disease monitoring and reporting system obtains suspected foodborne disease outbreaks by manually defining screening conditions, and the suspected outbreaks obtained by manual screening have a problem of low accuracy.

Images