As stated in the AI strategy announced by the Japanese government in 2019, “All high school graduates, regardless of their background, should acquire basic literacy in science, mathematics, data science, and AI” there is a very active movement to incorporate the basic acquisition process of such a content into school education. In actual courses, an educational environment is being created for students to acquire introductory knowledge of data science and programming skills necessary for practical use.
However, many barriers to learning data science can be expected for learners who regularly use various assistive technologies, such as braille displays and screen readers, when using computers. This is because there are still many issues that have not been well recognized, such as “visually dependent analysis methods unique to data science, such as analyzing large amounts of data graphically and from a bird’s eye view,” and “the environment for data science practice, which is still developing and inadequate in terms of accessibility. The reason is that there are still many issues that have not even been fully recognized yet.
Therefore, we have created these contents with the aim of providing tools and information to help people who are interested in learning data science but have difficulty learning due to accessibility issues.
Data science is an approach that aims to extract new insights from data, and it requires the utilization of interdisciplinary techniques such as information science, statistics, and algorithms to handle data. With the various trends of digitization, the importance of data science is increasing, and opportunities to learn and apply data science knowledge and skills are growing in various fields, including both businesses and schools regardless of their academic disciplines.
Exploratory data analysis refers to the process of examining and preprocessing data before conducting in-depth analysis in data science. It involves tasks such as confirming the given data and preparing it for easier handling, including visualizing data through graphs. For users relying on screen readers, data visualization can be one of the biggest barriers when learning data science. In the next section, we will introduce methods for auditory representation of data.
This content is primarily intended for students and professionals who use screen readers to operate computers, as well as teachers involved in the education of such students. While the practical application of data science involves Python programming, we also anticipate readers who are completely new to programming, so we will provide detailed explanations on setting up screen readers and other related aspects. However, we would like to focus on providing an accessible and approachable learning experience, and therefore, we will minimize detailed explanations of mathematics and programming. There may be unfamiliar terms that come up, but we encourage you to look them up on the internet or ask others for assistance as needed. It may be challenging at first, as there will be many unknowns and a lack of familiarity, but in programming and data science, the habit of “researching and resolving issues independently” is crucial. So please do your best and embrace this approach.
In this content, our goal is for you to become capable of independently exploring data science using Google Colab, an integrated development environment, while using a screen reader. The knowledge and techniques provided along the way are minimal but valuable for those who are interested in data science and wish to continue their exploration. The content is structured as follows, and in terms of volume, it can be completed in approximately one day by fast learners.
In the first chapter, you will learn about the fundamental technologies of screen readers and web browsers necessary for programming on Google Colab. Then, in the second chapter, you will learn the basics of Python programming, which is currently the most widely used programming language in data science. Next, in the third chapter, we will introduce methods of data representation using sound instead of visual graphs, as this is crucial for learning data science without relying on vision. Lastly, in the fourth chapter, you will apply the programming and graph utilization techniques learned in previous chapters to engage in introductory data science practice.