Welcome and Opening Remarks
Hiroki Sakaji, Teruaki Hayashi, Kiyoshi Izumi, and Yukio Ohsawa
Verification of Data Similarity using Metadata on a Data Exchange Platform
With the development of computers and the rise of data exchange, the expectations for innovation by combining data from different industries, and data exchange platforms that handle different types of data are sprouting up. However, the data handled on such platforms have been obtained and are stored independently by data providers with different purposes, and the maintenance of data catalogs and the spread of schemata are not currently insufficient, making it difficult to understand the relationships between the data on such platforms. In this study, in order to derive the relationships among datasets and to discuss the similarity and combinability of the data on the data exchange platform, we analyze and discuss the relationships between datasets on the platform. We focussed on the data outlines and variables using the metadata of the data exchange platform service, D-Ocean, and found that the similarity of data cannot be measured by a single indicator, and that the items to be referred to calculate the similarity depending on the indicator.
Teruaki Hayashi, Nao Uehara, Daisuke Hase, and Yukio Ohsawa
Data Requests and Scenarios for Data Design of Unobserved Events in Corona-related Confusion Using TEEDA
Due to the global violence of the novel coronavirus, various industries have been affected and the breakdown between systems has been apparent. To understand and overcome the phenomenon related to this unprecedented crisis caused by the coronavirus infectious disease (COVID-19), the importance of data exchange and sharing across fields has gained social attention. In this study, we use the interactive platform called treasuring every encounter of data affairs (TEEDA) to externalize data requests from data users, which is a tool to exchange not only the information on data that can be provided but also the call for data–what data users want and for what purpose. Further, we analyze the characteristics of missing data in the corona-related confusion stemming from both the data requests and the providable data obtained in the workshop. We also create three scenarios for the data design of unobserved events focusing on variables.
Mhd Irvan, Toshiyuki Nakata, and Rie Shigetomi Yamaguchi
User authentication based on smartphone application usage patterns through learning classifier systems
Smartphones have become more ubiquitous than ever. People are installing various applications on their smartphone to fit into their lifestyle. Existing research shows that there are patterns within the ways people access those applications, whether it involves particular locations, particular ranges of time, or many other factors. In this research, through a collaboration with a commercial company, we collected usage data from a popular smartphone application that gives its users access to digital flyers information for shops and supermarkets throughout Japan. Our early experiments found that the pattern information contained inside the data could be used to authenticate users. In this research, we are proposing a behavioral authentication model implementing customized learning classifier systems to search through vast amount of possible patterns to authenticate users of the application. Our early findings for this ongoing research demonstrate that our model can feasibly be a good alternative for additional authentication factor to implicitly authenticate users beyond the initial registration.
Bin Wang, Teruaki Hayashi, and Yukio Ohsawa
Hierarchical Graph Convolutional Network for Data Evaluation of Dynamic Graphs
As data is being generated at an incredible speed and scale, the market of data that aims to fully discover the value of data will become increasingly important. Data evaluation is a vital part of the data market. It can discover the data's flaws and the value hidden in the data. Nevertheless, how to conduct data evaluation in the data market has not been widely studied. Existing methods seldom utilize the multi-level structure in data. Our work proposes a novel hierarchical graph convolutional network for data evaluation of dynamic graphs, following the anomaly detection paradigm. Our model performs significantly better than existing models on several benchmark datasets. A more powerful tool for processing dynamic graphs is provided here. It could also guide the direction of data evaluation for a broader range of data categories in the data market.
Turning Big Data to Business advantage through Azure Analytics
Several years ago, needs of big data processing is a hot research topic in the big data era. Instead it forms the foundation for some of today's most exciting technologies. Artificial intelligence (AI), machine learning, and data science rely on big data, or data that—by virtue of its velocity, volume, or variety—cannot be easily stored or analyzed with traditional methods. If you want to know how you can thrive in the world of big data, data science, machine learning, and artificial intelligence you are at the right place. We will be focusing on Roles and capabilities of Microsoft Cloud in Big Data Domain. We will briefly be covering the Big data's relationship to AI, data science, social media, and the Internet of Things (IoT).
Uday Pathania is a Sr. Customer Engineer and spend 7 years in Microsoft. I have total 15 years of experience in Data domain under various Organization under different sector. From last couple of months I am mostly working with Unicorn customer’s across globe and helping them to grow/establish their businesses. Big data engineering and Artificial intelligence is my core strength.
General Comment and Closing Remarks
Dr. Teruaki Hayashi (co-chair)
Email: hayashi -at- sys.t.u-tokyo.ac.jp