Wolfram Computation Meets Knowledge

Wolfram Summer School


Naoki Sawahashi

Summer School

Class of 2015


Naoki Sawahashi is a computer science and math major from Tokyo, Japan, who hopes to develop a non-invasive interface between humans and computers in his future career.  He passed the Fundamental Information Engineer Examination, held by the ministry of the government of Japan (METI), when he was in the 11th grade.  He is currently working toward his bachelor degree at Hanover College in the United States, and expects to graduate in 2017. Naoki is a math tutor and a teacher assistant in the math department. He also is in charge of group tutoring sessions each week, teaches students one-on-one, and designs supplementary teaching sessions for Calculus 1 and Applied Statistics.

Naoki has a wide range of interests. He started his career as a classical pianist when he was four years old and gives classical recitals annually. He loves Chopin’s nocturnes and cinema music composed by Joe Hisaishi. He is also a serious photographer and videographer. Naoki does videography and photography for the communication department at the college. He will direct an educational video on math history in Italy next May and develop an educational website for the video. Naoki believes there is a certain relationship between artistic beauty and scientific integrity, and his belief has motivated him to study both science and arts throughout his entire life.

Project: Sonification of Time Series Data and Functions

My project is developing a system to represent various data as sound. The concept, showing data as nonverbal sound, also known as sonification, has not been commonly used in data analysis. The goal of my project is applying the concept of sonification to actual data analysis. As an example of the practical applications of sonification, this project produced a small website, Sonificator, which takes natural language expressions that describe specific time series data (e.g. “temperature in Boston, April 1994”) and automatically sonifies them.

The final result of this project is the alpha version of Sonificator, which currently has two methods to sonify the data in a web browser. The first method, Melody, creates a simple melody corresponding to the data points in a time series. It works with both small and large datasets by cutting off the relatively small change between data points throughout an entire time series. By contrast, Frequency Mapping describes the change of data by dynamically changing the frequency of the sine wave. Therefore, the generated sound keeps many more details as compared with the Melody method. However, it is sometimes very hard to follow the sound produced by Frequency Mapping because the frequency changes dramatically in a very short time period. The next step of the project is to implement a function that chooses the proper method automatically for different sized datasets.