As the Internet develops, the number of Internet of Things (IoT) devices increases. Simultaneously, the risk of IoT devices being infected with malware also increases. Thus, malware detection has become an important issue. Dynamic analysis logs are effective at detecting malware, but it takes time to collect a large amount of data because the malware must be executed at least once before the logs can be collected. Moreover, dynamic analysis logs are affected by external factors such as the execution environment. A malware detection method that uses a static property analysis log could solve these problems. In this study, deep learning (DL) was used as a machine learning method because DL is effective for large-scale data and can automatically extract features. Research has been conducted on malware detection using static properties of portable executable (PE) files, establishing that such detection is possible. However, research on malware detection using hash functions such as Fuzzy hash and peHash is lacking. Therefore, we investigated the characteristics of hash values in malware classification. Moreover, when the surface analysis log is viewed in chronological order, that the data are considered have concept drift characteristics. Therefore, we compared malware detection performance using data with the concept drift property. We found that the hash function could be used to prevent performance degradation even with concept drift data. In an experiment combining PE surface information and hash values, concept drift showed the highest performance for certain data.

In 2020, the Government of Japan declared "2050 carbon neutral" and launched a long-term strategy to create a "virtuous cycle of economy and environment".(1) Japanese firms possess many technologies that contribute to decarbonization, which is important to expand investment for Green Technology (environmental technology) development. As automobiles are major contributors to greenhouse gas emissions [1], the technological shift towards vehicle powertrain systems is an attempt to lower problems like emissions of carbon dioxide, nitrogen oxides [2]. On the other hand, patent data are the most reliable business performance for applied research and development activities when investigating the knowledge domains or the technology evolution (Wand, 1997). Our paper describes a Japanese patents dataset of the vehicle powertrain systems for hybrid electric vehicle (HEV), battery electric vehicle (BEV) and fuel cell electric vehicles (FCEV). In this paper we create a method of bombinating international patent classification (IPC) and keywords to define "green" patents in vehicle powertrains field, using patent data which were applied to Japan Patent Office recorded on EPO's PATSTAT database during 2010 similar to 2019 year. When analyze patents, it is necessary to consider the social situation of each country including language background, we collect patents description documents (abstracts and titles) not only written in English but also in Japanese. Finally, we build a database includes 6025 green patents' description documents and 266 patents' holding firms. With which we then identify 3756 HEV patents, 1716 BEV patents, and 553 FCEV patents. Data about patent holding firms is also appended. The full dataset may be useful to researchers who would like to do further search like natural language processing and machine learning on patent description documents, statistical data analysis for empirical economics. (c) 2022 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/)

Mental health illness has become a serious public problem. Finding changes in everyday behavior is a demand. This paper tries to detect persons who have mental health illness using their short comments posted to social network systems. The novelty of this study is using comments in a system for communication between users with mental health illness, in order to prepare a sufficient amount of supervised data for machine learning. The authors used approximately 120,000 comments in the system as positive samples and 120,000 comments in Twitter as negative samples for detecting mental health illness. Both data are posted short comments on a daily basis. The authors conducted a straightforward classification of the comments using a support vector machine and surface-level features of the comments. The accuracy of the classification is 0.92 and the characteristic phrases used for the classification are related to troubles in mental health. The ability to classify everyday statements can be expected to lead to the early detection of mental disorders.

Researchers are expected to find previous literature that is related to their research and potentially has a scientific impact from among a large number of publications. This paper addresses the problem of predicting the citation count of each research paper, that is, the number of citations from other papers to that paper. Previous literature related to the problem claims that the textual data of papers do not deeply affect the prediction compared with data about the authors and venues of publication. In contrast, the authors of this paper detected the citation counts of papers using only the paper abstracts. Additionally, they investigated the effect of technical terms used in the abstracts on the detection. They classified abstracts of papers with high and low citation counts and applied the classification to the abstracts modified by hiding the technical terms used in them. The results of their experiments indicate that the high and low of citation counts of research papers can be detected using their abstracts, and the effective features used in the prediction are related to the trend of research topics.

Predicting author’s attributes is useful for understanding implicit meanings of documents. The target problem of this paper is predicting author’s native language for each document. The authors of this paper used surface-level features of documents for the problem and tried to clarify the practical tendencies of the writing style as word occurrences. They conducted a classification of the abstracts written in English of approximately 85,000 scholarly papers written in English or in Japanese. As a result of the experiment, the accuracy of the binary classification was 0.97, and they found that a number of distinctive phrases used in the classification were related to typical writing styles of Japanese.

Efficient methods are required for plagiarism detection. This paper proposes a fast and scalable method for detecting "copy and paste"-type plagiarism in documents. Implementing detection methods for this type of plagiarism requires a long processing time or a large database for comprehensive matching of ordered word occurrences. The author improved the scalability of an existing fast method based on fast Fourier transform using the idea of the frequency domain filtering. He evaluated the effect of the improvement on accuracy of the plagiarism detection method, and achieved an effective trade-off between the accuracy and the required size of database.

An institutional repository (IR) is a system for archiving digital research outputs such as scholarly papers. While using this system is a simple way to return the benefits of academic research to society, the number of papers currently archived in institutional repositories across the world is very small. We have, therefore, developed a system that can increase the number of papers archived in institutional repositories. The main aim of this system is to reduce the workload of researchers and librarians in paper registration for an IR by using external databases. As a result of our trial implementation and evaluation of the system, we found that its application resulted in a fourfold increase in the number of papers registered in a month without decreasing the working efficiency of the participating librarians. Moreover, by analyzing the log data for the system's implementation, we have acquired further knowledge related to researchers who need librarians' support in paper registration with an IR. © 2018 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc.

Researchers are required to find previous literature which is related to their research and has a scientific impact efficiently from a large number of publications. The target problem of this paper is predicting the citation count of each scholarly paper, that is, the number of citations from other scholarly papers, as the scientific impact. The authors tried to detect the high and low of the citation count of scholarly papers using only their abstracts, especially, non-technical terms used in them. They conducted a classification of abstracts of scholarly papers with high and low citation counts, and applied the classification also to the abstracts modified by deleting technical terms from them. The results of their experiments indicate that the scientific impact of a scholarly paper can be detected from information which is written in its abstract and is not related to the trend of research topics. The classification accuracy for detecting scholarly papers with the top or bottom 1% citation counts was 0.93, and that using the abstracts without technical terms was 0.90.

Plagiarism detection for a huge amount of document data requires efficient methods. This paper proposes a plagiarism detection algorithm based on approximate string matching and vector representation of words, and a speed improvement to an implementation of the algorithm. The effect of the improvement on the algorithm is evaluated by conducting experiments with a dataset. The experimental results show a tradeoff between the processing time and the accuracy of the plagiarism detection algorithm, which enable us to configure its implementation in accordance with a given data space and a required accuracy.

The match-count problem on strings is a problem of counting the matches of characters for every possible gap of the starting positions between two strings. This problem for strings of lengths m and n (m≤n) over an alphabet of size σ is classically solved in O(σnlog⁡m) time using the algorithm based on the convolution theorem and a fast Fourier transform (FFT). This paper provides a method to reduce the number of computations of the FFT required in the FFT-based algorithm. The algorithm obtained by the proposed method still needs O(σnlog⁡m) time, but the number of required FFT computations is reduced from 3σ to 2σ+1. This practical improvement of the processing time is also applicable to other algorithms based on the convolution theorem, including algorithms for the weighted version of the match-count problem.

In recent years, the more popular automatic identification systems using RFID have become, the more important privacy problems have become, with the phrase such as, “It is necessary to regard privacy for RFID’s success”. At the same time, not many engineers, users, and RFID tag vendors, understand RFID privacy problem comprehensively, unfortunately. RFID is considered to penetrate wide application areas, some of which might be highly sensitive to users’ privacy. Although applications in such privacy-sensitive areas have been avoided such as RFID tags are “killed” before handed to consumers with products, there are also predictions that RFID’s power exists in the areas which involve consumers, that is, users. We, being involved in RFID, must advance our knowledge and understanding of RFID privacy to manage troubles regarding users’ privacy properly and quickly. In this chapter, we try to clarify, as technical as possible, problems with respect to privacy and personal information protection around RFID systems.

This paper proposes a fast plagiarism detection algorithm in large-scale data. Plagiarisms of superficial descriptions, such as “copy and paste”, can be detected using a simple document similarity based on string matching. The algorithm reduces the effort for computing the document similarity by approximating the similarity. The effects of the approximation on the processing time and accuracy are evaluated by conducting experiments with a data set generated from practical scholarly documents. The experimental results show that the algorithm based on the approximated similarity can reduce the processing time of the straightforward algorithm based on the exact similarity to less than one-third in exchange for a slight decrease of the accuracy. Subject Categories and Descriptors H.3.3 [Information Search and Retrieval]; H.2.4 [Systems]. General Terms: Plagiarism Detection, String Matching, Large Datasets

It is often pointed out that students' academic performance becomes worse. Lack of professors' teaching ability is often considered its major cause, and universities promote faculty development programs. According to our observation, however, the major cause is rather on student's side, such as lack of motivation, diligence, and other attitudes toward learning. In this paper, we focus on diligence. Diligence is quite important for students to learn effectively. Among various kinds of diligence, we take two kinds of them into consideration; the length of answer text to a questionnaire, and the amount of submitted homework assignments. We investigate how these kinds of diligence of students relate each other, and how they relate to the examination score.

Accurate methods are required for plagiarism detection from documents. Generally, plagiarism detection is implemented on the basis of similarity between documents. This paper evaluates the validity of using distributed representation of words for defining a document similarity. This paper proposes a plagiarism detection method based on the local maximal value of the length of the longest common subsequence (LCS) with the weight defined by a distributed representation. The proposed method and other two straightforward methods, which are based on the simple length of LCS and the local maximal value of LCS with no weight, are applied to the dataset of a plagiarism detection competition. The experimental results show that the proposed method is useful in the applications that need a strict detection of complex plagiarisms.

The match‐count problem on strings is the basic problem of counting the matches of characters between two strings for every possible alignment. The problem is classically computed in O(σ n log m) time using a fast Fourier transform (FFT) for two strings of lengths m and n (m ≤ n) over an alphabet of size σ. This paper extends the target of this FFT‐based algorithm to a weighted version of the problem, which computes the sum of similarities between characters instead of the number of matches. The algorithm extended in this paper can solve the weighted match‐count problem in O(dn log m) time by mapping characters to numerical vectors of dimensionality d. This paper also evaluates the usefulness of the extended algorithm by applying it to plagiarism detection in documents. The experimental results show that the proposed algorithm is applicable to general vector representation of words and that the obtained plagiarism detection method can extremely reduce the processing time with a slight decrease of accuracy from the method based on the normal match‐count problem.

Plagiarism detection in documents requires appropriate definition of document similarity and efficient computation of the similarity. This paper evaluates the validity of using vector representation of words for defining a document similarity in terms of the processing time and the accuracy in plagiarism detection. This paper proposes a plagiarism detection algorithm based on the score vector weighted by vector representation of words. The score vector between two documents represents the number of matches between corresponding words for every possible gap of the starting positions of the documents. The vector and its weighted version can be computed efficiently using convolutions. In this paper, two types of vector representation of words, that is, randomly generated vectors and a distributed representation generated by a neural network-based method from training data, are evaluated with the proposed algorithm. The experimental results show that using the weighted score vector instead of the normal one for the algorithm can reduce the processing time with a slight decrease of the accuracy, and that randomly generated vector representation is more suitable for the algorithm than the distributed representation in the sense of a tradeoff between the processing time and the accuracy.

Plagiarism detection in a large number of documents requires efficient methods. This paper proposes a plagiarism detection algorithm based on approximate string matching to be specified in "copy and paste"-type plagiarisms, and a speed improvement to an implementation of the algorithm. Most of the computations required in the algorithm are omitted by two kinds of approximations of the output used for plagiarism detection, while the decrease of accuracy caused by the approximations is acceptable. The effect of the improvement on the processing time and accuracy of the algorithm is evaluated by conducting experiments with a data set. The experimental results show that the improvement can reduce the processing time to approximately one-twentieth for a 6.4% decrease of the accuracy from those for the normal implementation of the algorithm.

Due to the advancement of popularization of university education, it becomes more and more necessary for university staff to help students by enhancing their motivations to learn in addition to training study skills. We approach to this problem from lecture data analytics. We have been investigating students’ answer to a term-end retrospective questionnaire, and found students’ attitude in learning and their academic performance correlate significantly. On the basis of this finding, in this paper, we propose a framework for assisting students to improve their learning attitude. It consists of four participants; lecturer, assisting staff including librarian, data analysts, and learning assistance system built on top of learning management system. We discuss how the results of our previous studies can be utilized to assist students in this framework. Further, we introduce two indexes for measuring the weights of a student viewpoint between lecture and themselves, and between good points and bad points. These indexes show how a student’s viewpoint to the class is located in comparison with other students’ viewpoints.

Understanding the states or emotions of learners at a lecture is expected to be useful for improving lecture quality. In our work, we tried to recognize two activities of learners by using their brain wave data to estimate their states. While existing analyses of brain wave data for activity recognition used standard bands such as α and β as features, we used other bands with higher and lower frequencies to compensate for the coarseness of simple electroencephalographs. We conducted experiments on recognizing two activities performed by six subjects with brain wave data captured by a simple electroencephalograph. We applied a support vector machine to 8-dimensional vectors corresponding to eight bands of the brain wave data. The results show that using the eight bands yielded higher accuracy compared than that obtained with the standard features based on at most four bands.