The purpose of this study is to support reviewing on their own decision for foreign exchange trends. Foreign exchange trends can be predicted from economic and political events. This is a decision based on experience, and it enables improvement of accuracy by verifying the validity of own analysis. However, as the factors of foreign exchange fluctuations are diverse, it is difficult to verify the adequacy of foreign exchange trends.To solve this problem, this paper proposes a method to classify news articles based on investor analysis factor in order to grasp the relevance of news articles and foreign exchange trends.

This paper describes an interpretation support system for classification patterns based on the contents of learning results in deep learning with texts, and verified its effectiveness. It is well known that classification patterns by deep learning models are often difficult to interpret the reasons derived. The proposed system extracts the contents of learning results in deep learning with texts and provides seeds for interpretations of the patterns learned. Then, the system displays learned network structures so that anyone can easily understand learning results. In verification experiments to confirm the effectiveness of the system, based on the learning result of deep learning classifying sentences, test subjects were instructed to give meanings of classification patterns peculiar to each output. The results show that test subjects who represent novice data scientists could understand the meanings of the classification patterns of deep learning with texts.

In recent years, many scientific indicators have been proposed for measuring the impact of research, and as the selection criteria of academic journals. Although these indicators give us quantitative evaluation based on the number of citations and downloads of papers, and the number of publications of authors, there is a problem that these indicators are not based on the contents of a paper. In order to support researchers to find appropriate journals for their paper submission and information gathering, a novel indicator that overcomes the above-mentioned problem is needed. This paper compares existing scientific indicators as a preliminary stage for the final objective of research evaluation based on contents

The 41st Annual International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR 2018) took place in Ann Arbor, USA in July 2018. A total of 409 papers for the full paper track were submitted and reviewed by six tracks, for an acceptance rate of 21%. About 46.0% of the accepted papers are related to neural network, distributed over about 73.1% of technical sessions.