Intern Blog by Mr. Ken Honda / インターンブログ(本多 賢さん)

My name is Ken Honda, and I've been working at OrangeTechLab (OTL) as an intern for about 2 years since June 2018. So, in this blog, I'd like to reflect on my encounter with OTL and its activities!

The reason I emphasized "as an intern" is that I had a relationship with the OTL students before I started my internship at the company.

In 2016, a project for research and practice was launched at Komazawa University, where I was enrolled. Professor Yoshida of my laboratory was involved in the launch of this project, and he asked me to participate in the project. At that time, I met the current president of OTL Miyazaki and other members of the project, and I was able to participate in the regular meetings of the project. At that time, OTL hadn't been established yet, so it was a great experience for me to see the company being built up close and personal.

When I first joined, I was still a graduate student in a master's program, and I remember that the regular meetings of the project were always exciting and I was always nervous. It was my best effort to keep up with the talk because I had changed my mind to the information system when I went on to the graduate school from the undergraduate school, and I still had neither enough knowledge nor experience. Since it was a place where professionals gathered, I learned a lot, and I think I was able to set the standard for my current knowledge, skills and experience.

In 2017, I started my PhD program at the same time that OrangeTechLab was founded, and I will be interning at OTL. I was involved in various activities, mainly in survey and reverse engineering. I am not involved in development, but I was interested in machine learning and deep learning because I saw products and projects made by deep learning with my own eyes while working in OTL. My research is aimed at solving the theme of social psychology, "Fake News", by using information technology such as databases (I call it credibility evaluation technology). In my paper, I mention AI and machine learning, and I use natural language processing as a technology. These are the ideas and techniques that I have adopted based on what I have noticed and learned through my activities in OTL.

This figure shows the main technical elements related to the current credibility assessment techniques. Although I mention it a little more in the paper, there is a lot of material from OTL's activities in this study. I was very grateful and warmed by the fact that she cared not only about my work, but also about my graduate school life and asked me to do work that would be useful for my research.

The most memorable part of the OTL activity was the experience of surveying a technology, delving into it, summarizing it, and presenting it in front of a client. It was the first time for me to do everything here. You can't miss! And I was fixing the materials until just before. It was a very different experience from the research and it was exciting.

I learned not only about research and technology, but also about how to proceed with projects and what I need to do as a member of the workforce, which are common skills and experiences for me now. I think it is only because of the activities at OTL that I was able to complete the doctoral course in three years and obtain a doctorate. This is really not an exaggeration, but the actual activities that we have done at OTL, such as the survey we did, are cited in our doctoral dissertation, so it's really true!

I'm looking forward to the opportunity to improve my skills in the field and to work with others in OrangeTechLab.

Thank you so much!

Translated with (free version)


OrangeTechLab (以下OTL) でインターンをしていた本多 賢です!2018年の6月から2年弱ほど「インターンとして」お世話になりました.そこでこのブログでは OTL との出会いや活動について振り返っていきたいと思います!


2016年,私が在籍していた駒澤大学に研究と実践のためのプロジェクトが立ち上がりました.このプロジェクトの立ち上げに,私が在籍する研究室の吉田教授が関わっていたということもありプロジェクトに参加しないかとお声がけいただいたのが最初の縁でした.この時に出会ったのが現在のOTL宮崎社長やメンバーの方々で,プロジェクトの定例ミーティングに参加させてもらっていました.当時はまだ OTL も設立されていなかったので会社が作られていく様子も間近で見ることができて,とてもいい経験でした.


2017年に OrangeTechLab が設立されると同時に博士課程に進学し,OTLにはインターンとして参加することになりました.サーベイやリバースエンジニアリングなどを中心に様々な活動に携わりました.開発という開発にはほんのちょっぴりしか携わってはいないのですが,機械学習や深層学習に興味を持ったのもOTLで活動する中で実際に深層学習で作られたプロダクトやプロジェクトを自分の目で見た,というのが大きかったです. 私の研究は社会心理学のテーマである「流言・フェイクニュース」をデータベースなどの情報技術で解決することを目指すものです (信憑性評価技術と呼んでいます).論文内では AI や機械学習にも触れていますし,技術としては自然言語処理などを使っています.これらは OTL で活動してきたことがきっかけで気づいたことや知ったことを元に取り入れた考えや技術です.

この図は現在の信憑性評価技術に関係する主な技術要素を示したものです.論文の中ではもう少し触れていますが,この調査に関しては OTL の活動から得たものがとても多く含まれています.業務だけではなく,大学院生活のことも気にかけていただき,せっかくだから自分の研究に役立つ業務をしてほしい,と寄り添っていただいたのがとてもありがたく,温かかったです.

OTL での活動で印象に残っているものはサーベイとして1つの技術を掘り下げ,まとめ,クライアントの前で発表する,という経験です.ここでも全てが初めて尽くしでした.ミスれない!と直前まで資料を直したりしていました.研究とは全く違う経験で刺激的でした.

研究や技術の事だけでなく,プロジェクトの進め方や,そもそもの社会人として必要な事など 今の自分にとって当たり前であるスキルや経験を OrangeTechLab との関わりから学ぶことができました.無事に博士課程を3年で修了し博士号を取得することができたのも OTL での活動があったからこそだと思っています.これは本当に大袈裟などではなく,実際に OTL で行った活動,例えばサーベイで調べたことなどは博士論文の中で引用されているので本当にその通りなのです!

OrangeTechLab での活動を通して学んだことを活かし今後も同じ分野で活動していくことになります,社会に出てレベルアップし,またOrangeTechLab のみなさんとお仕事ができる機会を楽しみにしています.