Twitter is currently one of the most influential microblogging services on which users interact with messages. It is imperative to grasp the big picture of Twitter through analyzing its huge stream data. In this study, we develop a two-stage clustering method that automatically discovers coarse-grained topics from Twitter data. In the first stage, we use graph clustering to extract micro-clusters from the word co-occurrence graph. All the tweets in a micro-cluster share a fine-grained topic. We then obtain the time series of each micro-cluster by counting the number of tweets posted in a time window. In the second stage, we use time series clustering to identify the clusters corresponding to coarse-grained topics. We evaluate the computational efficacy of the proposed method and demonstrate its systematic improvement in scalability as the data volume increases. Next, we apply the proposed method to large-scale Twitter data (26 million tweets) about the COVID-19 Vaccination in Japan. The proposed method separately identifies the reactions to news and the reactions to tweets.

In this paper, we compare the three countries, India, Japan, and Indonesia's Twitter topics concerning COVID-19. The tweet data were collected from the period of April 2021 to June 2021. The damage of COVID-19 in India and Indonesia was unprecedented in our human beings history. Unexpectedly we could collect Tweets concerning the pandemic. From the data, we would like to extract unexpected topics to prepare for future challenges from humanitarian standpoints. In Japan, the female suicide rate raised significantly. In India, the Joint Entrance Examination were cancelled due to the pandemic, which caused irregular educational systems timeline for Indian students which might create a lot of concerns in future. In Indonesia, Tweets during the peak period on June 2021 have shown some record of discussion on the scarcity of oxygen tubes at isolation houses during the surge of COVID-19 pandemic cases. The results of the analysis showed us the growing fear of the infection, the situation of lack of oxygen tubes, the sad news of the increase in suicides in Japan, the confusion of the entrance examination system nationwide in India, and services related to the distribution of subsidies from the government in Indonesia.

Abstract During a disaster, social media can be both a source of help and of danger: Social media has a potential to diffuse rumors, and officials involved in disaster mitigation must react quickly to the spread of rumor on social media. In this paper, we investigate how topic diversity (i.e., homogeneity of opinions in a topic) depends on the truthfulness of a topic (whether it is a rumor or a non-rumor) and how the topic diversity changes in time after a disaster. To do so, we develop a method for quantifying the topic diversity of the tweet data based on text content. The proposed method is based on clustering a tweet graph using Data polishing that automatically determines the number of subtopics. We perform a case study of tweets posted after the East Japan Great Earthquake on March 11, 2011. We find that rumor topics exhibit more homogeneity of opinions in a topic during diffusion than non-rumor topics. Furthermore, we evaluate the performance of our method and demonstrate its improvement on the runtime for data processing over existing methods.

This paper presents FITS, or Feature-value / Instance Transposition Selection, a method for unsupervised clustering. FITS is a tractable, explicable clustering method, which leverages the unsupervised feature value selection algorithm known as UFVS in the literature. FITS combines repeated rounds of UFVS with alternating steps of matrix transposition to produce a set of homogenous clusters that describe data well. By repeatedly swapping the role of feature and instance and applying the same selection process to them, FITS leverages UFVS's speed and can perform clustering in our experiments in tens milliseconds for datasets of thousands of features and thousands of instances.We performed feature selection-based clustering on two real-world data sets. One is aimed at topic extraction from Twitter data, and the other is aimed at gaining awareness of energy conservation from time-series power consumption data. This study also proposes a novel method based on iterative feature extraction and transposition. The effectiveness of this method is shown in an application of Twitter data analysis. On the other hand, a more straightforward use of feature selection is adopted in the application of time series power consumption data analysis.

With the increasing use of smartphones, lots of smartphone based applications have been developed. Smart-phones are used in personal health care or monitoring activities of elderly persons. These types of smartphone applications require continuous authentication of the user for taking action in case of detachment of the smartphone from the user due to forgetfulness or theft. Continuous authentication on smartphone requires authentication process having low computational overhead. In this work, the objective is to develop low cost user authentication algorithm from time series data of user activities taken from sensors like accelerometer or gyroscope. Deep neural networks are used for user authentication. A two-step authentication process has been developed in which sensor data has been first classified into different activities and activity dependent authentication is proposed. For lowering computational cost of classifier, knowledge distillation is used to reduce the model parameters. Fine tuning is used to cope with the limited number of training data. As a result the authentication accuracy has been improved by 5% to 10%, also authentication time of 0.032 sec has been achieved which is useful for real time authentication. Simulation studies have been done by several bench mark data sets to evaluate the efficiency of the proposed approach.

In this paper, we contrast three regression methods on the example of finding the correlates of child mortality rates (under5mort) by province in Indonesia. Factors examined include the average high-school enrollment rate (enrollment), average level of poverty (poverty), and the total fertility rate (TFR). The three methods we compared are: 1) traditional multiple linear regression (MLR), 2) eXtreme Gradient Boosting (XGBoost) algorithm, and 3) Random Forest (RF) algorithm. We find that TFR and poverty show statistically significant relationship with the mortality rates of children under age 5, while the high-school enrollment rate does not. Results are qualitatively same for all three methods but differ in the value of coefficient of determination R-squared. XGBoost has highest coefficient of determination and is thus the best fitting model for our data samples. Based on order of features, the most important among the three factors is TFR and the next is poverty level. We propose that XGBoost and RF regression methods could be successfully applied to other demographic analyses in which the relationship between the feature variables and the predictor is not linear.

Recently, Cool Japan, expresses modern Japanese culture such as comic, fashion, technology, food and sightseeing, is getting popular all over the world. Especially, Japanese comic and fashion gained great popularity in Asian countries as one of the major contents of Cool Japan. This research proposes the social information platform of Cool Japan that targets Asian countries’ people for exploring the potential market. Our platform is based on a network consisting of universities in Asia that uses the social media to collect people’s thoughts and reactions. As the first step, we target Japanese fashion and comic, analyze people’s sensitivity using natural language processing techniques and our original feature selection technique. Our platform aims to collect the social information about Cool Japan contents in Asia by exploring the potential markets.

This paper is showing a time series topic life cycle extraction from millions of Tweets using our original community detection technique in bipartite networks. We suppose that the authors role that means who belong to what topics is important to extract quality topics from social media data. We already proposed the topic extraction method that considers the relationship between the authors and the words as bipartite networks and explores the authors role by forming clusters as topics. As the next step, this paper applies our method to the time series topic life cycle detection. We extract topics in different time slots and analyze the time series of topic transition using the coherence measure that expresses the semantic accuracy of topics. The paper demonstrates that our method can detect the topic life cycle such as the growth, the conflicts and so on over time from millions of Tweets.

It has been very significant to visualize time series big data. In the paper we shall discuss design on time series statistical data. As an example, we present an animation of Gibbs sampling process to clarify the time changes. Gibbs sampling is widely used MCMC algorithm in the deep learning field. We consider that an additional z-axis coordinate or a time line are helpful for the visualization and the functions could be implemented automatically by some kind of chart-wizards. We shall discuss the design rules and tips on visualization of time series data.

Social media offers a wealth of insight into how significant topics such as the Great East Japan Earthquake, the Arab Spring, and the Boston Bombing affect individuals. The scale of available data, however, can be intimidating: during the Great East Japan Earthquake, over 8 million tweets were sent each day from Japan alone. Conventional word vector-based topic-detection techniques for social media that use Latent Semantic Analysis, Latent Dirichlet Allocation, or graph community detection often cannot extract appropriate topics from such a large volume of data with accuracy due to their space and time complexity. To alleviate this problem, we propose an effective topic extraction from millions of tweets based on community detection in bipartite networks. Our method is based on the bipartite community detection technique developed by Okamoto, one of the authors of this paper. The paper demonstrates our method effectiveness on social media analysis and identifies topics from millions of tweets after the Great East Japan Earthquake. To show our method's effectiveness, we compute the coherence measure that can evaluate the semantic accuracy and the running time, and compare the method with LDA that is the major topic model.

Feature selection is a useful tool for identifying which features, or attributes, of a dataset cause or explain phenomena, and improving the efficiency and accuracy of learning algorithms for discovering such phenomena. Consequently, feature selection has been studied intensively in machine learning research. However, advanced feature selection algorithms that can avoid redundant selection of features and can detect interacting features require heavy computation in general and hence are seldom used for big data analysis. To eliminate this limitation, we tried to improve the run-time performance of two of the most advanced feature selection algorithms known in the literature. We have developed two accurate and extremely fast algorithms, namely Super CWC and Super LCC. In experiments with multiple real datasets which are actually studied in big data research, we have demonstrated that our algorithms improve the performance of their original algorithms remarkably. For example, for two datasets, one with 15,568 instances and 15,741 features and another with 200,569 instances and 99,672 features, Super-CWC performed feature selection in 1.4 seconds and in 405 seconds, respectively. This is a remarkable improvement, because it is estimated that the original algorithms would need several hours to a few ten days to perform feature selection on the same datasets.

In this paper, the minutes of the monetary policy of the Bank of Japan, the central bank of Japan, for the year 1998 has been analyzed for the extraction of the topics concerning Asian financial crisis. The currency crisis started in Thailand in 1997 and spread toward many Asian countries. We analyzed the Monetary Policy Meeting minutes by text mining technologies. Especially we conducted topic extraction from the meeting minutes using Latent Dirichlet Allocation (LDA) model and the time series of the changes of extracted topic ratios are shown. From the analysis results, one topic which seems to be the Asian financial crisis related topic can be found. The topic ratio change curve clearly corresponds to the calm down of economic indices such as currency exchange ratios and market interest rates in the Asian countries.

After the East Japan Great Earthquake occurred in Japan on 11th March 2011, a large number of messages related to the earthquake were posted to social media web sites not only in Japan but also all over the world. Particularly, in Asian countries, a lot of topics concerned about the earthquake were observed. Exploring topics related to the earthquake on social media gains rich insights into the social contexts. The goal of this research is to analyze Asian people reactions to the East Japan Great Earthquake on social media. As the first target, this paper selected Thai language and analyzed how people reacted to the earthquake by comparing with reactions in Japan. This analysis also presented some characteristics of Thai society and culture.

This paper proposes a time series topic extraction method to investigate the transitions of people's needs after the East Japan Great Earthquake using latent semantic analysis. Our target data is a blog about afflicted people's needs provided by a non-profit organization in Tohoku, Japan. The method crawls blog messages, extracts terms, and forms document-term matrix over time. Then, the method adopts the latent semantic analysis and extract hidden topics (people's needs) over time. In our previous work, we already proposed the graph-based topic extraction method using the modularity measure. Our previous method could visualize topic structure transition, but could not extract clear topics. In this paper, to show the effectiveness of our proposed method, we provide the experimental results, and compare them with our previous method's results.

Once a disaster occurs, people discuss various topics in social media such as electronic bulletin boards, SNSs and video services, and their decision-making tends to be affected by discussions in social media. Under the circumstance, a mechanism to detect topics in social media has become important. This paper targets the East Japan Great Earthquake, and proposes a time series topic detection method based on modularity measure which shows the quality of a division of a network into modules or communities. Our proposed method clarifies emerging topics from social media messages by computing the modularity and analyzing them over time, and visualizes topic structures. An experimental result by actual social media data about the East Japan Great Earthquake is also shown.

Knowledge management has been a hot research issues for decades and resulted in many proposals and some systems that have been supporting information exchange in companies or societies. They failed however to become knowledge systems. With the advent of the internet the knowledge web is going to replace knowledge management. It creates its own challenges but take advantage of all web, information systems and software technology.

This paper proposes a method to discover consumer behavior from buzz marketing sites. For example, in 2009, the super-flu virus spawned significant effects on various product marketing domains around the globe. Using text mining technology, we found a relationship between the flu pandemic and the reluctance of consumers to buy digital single-lens reflex camera. We could easily expect more air purifiers to be sold due to flu pandemic. However, the reluctance to buy digital single-lens reflex cameras because of the flu is not something we would have expected. This paper applies text mining techniques to analyze expected and unexpected consumer behavior caused by a current topic like the flu. The unforeseen relationship between a current topic and products is modeled and visualized using a directed graph that shows implicit knowledge. Consumer behavior is further analyzed based on the time series variation of directed graph structures.

This paper presents a concept model for solving bond mathematics problems; it is the manner in which our knowledge concerning bond mathematics is organized into a cognitive architecture. We build a concept model for bond mathematics to achieve good organization and to integrate the knowledge for bond mathematics. The ultimate goal is to enable many students to understand the solution process without difficulty. Our concept model comprises entity-relationship diagrams, and using our concept models, students can integrate financial theories and mathematical formulas. This paper illustrates concept models for bond mathematics by showing concrete examples of word mathematics problems. It also describes our principles in developing the concept model and the descriptive power of our model.

A huge amount of multimedia content is available in the current Web environment. Streaming of video programs has become one of the popular Web-based information services. Automatic digest generation is an important application using video program streams. We have developed a digest making method called PDMS (Personal Digest Making Scheme). PDMS extracts significant scenes and constructs digests automatically using the video program meta data. In PDMS, only a successful play event such as a goal scored was considered significant, and a misplay event, such as a shot that did not score was not taken into account, although such an event may affect the game progress. This paper introduces a new concept of turning point analysis into PDMS. The turning point analysis is based on a winning probability for a sports programs. The winning probability indicates the probability of a home team beating an away team at the end of the game, given the current score and the time elapsed since the beginning of the game. Using the winning probability, we can more precisely evaluate the significance of each event, not merely a successful play but also a misplay. This paper presents turning point analysis for a soccer match. It also gives evaluation results of this extended PDMS, including turning point analysis for a recently broadcasted professional soccer match.