A News Recommender System for Media Monitoring

摘要: Media monitoring services allow their customers, mostly companies, to receive, on a daily basis, list of documents from mass media that discuss topics relevant the company. However, often generate these lists by using keyword-filtering techniques, which introduce many false positives. Hence, before end users, i.e., employees company, may consult and find documents, human editor must inspect keyword-filtered remove This is time consuming job. In this paper we present recommender system aims at reducing number needs every day. The proposed solution classifies (represented with TF-IDF embeddings features) techniques trained data containing editors’ past actions (i.e. removals positives). technique shown be able correctly predict true positives, thus