Data Vehicle, Inc.

Big data analyzation system

Takram, in collaboration with Dentsu CDC, was involved in the design, planning, and development of “DataDiver,” a data analysis tool that was released from Data Vehicle Inc. in 2015. Until now, tools for collecting and analyzing data had been designed for expert users of the field with advanced analysis skills and experience. However, “DataDiver” is designed for users without such expertise, and can lead anyone to find important information that can be used in the decision-making process. In this project, Takram designed a tool that makes analysis easy and simple by focusing on intuitive infographics and a pleasant user experience.

CONCEPT: Combining simple operation with full-scale analysis

Data analysis is usually performed in a number of steps, but in “DataDiver” the process is simplified to focus only on the significant activities by automating the parts that are less important. This enables “DataDiver” to combine simple operability with an ability to perform full-scale data analysis. The “less is more” user experience defines the entire service as the project’s core design concept.

Furthermore, big data has traditionally been used for hypothesizing based on conclusions derived from its analysis. However, the process hasn’t been enough to make big-picture decisions regarding the entire data set. The ability to weave the two axes of analysis and visualization, and zoom in and out of the detail to the whole, are also part of the “DataDiver” concept.

DETAIL: User operability

User operations in “DataDiver” are very simple. Data analysis and visualization are completed by following these three steps: setting up the purpose for analysis, confirming the provided analysis, and creating a report. In confirming the results of the analysis, various graphics and animation help the user to properly understand the information. The various processes besides these three steps, such as estimating the relationships between data tables, processing gathered data, natural language processing, and graph and animation visualization are algorithmically automated to enable non-expert users to perform full-scale analysis on “DataDiver.”

BACKGROUND: Analyzing data for business strategies

In the manufacturing field today, vast amounts of data are being collected and analyzed for purposes of product quality control, new product planning, and business strategies. On the other hand, detailed studies of such information require the use of data analysis tools, which are usually designed for expert and skilled analysts. Furthermore, in many cases, the preparation and actual analyses are performed by in-house data scientists. The interpretation and decision-making based on these analyses are done by managers and are implemented by the person in charge of the actual operation. This cumbersome process is inefficient, and more often than not, there aren’t enough data scientists working on the field to start with. Easy operation and simple graphics enable more people to analyze data, empowering them to quickly arrive at executable and realistic solutions based on the program’s highly accurate hypotheses.