Shunsuke Kanda (Ph.D.)
This is a personal website of Shunsuke Kanda (神田峻介) or Kampersanda (かんぱさんだ).
About me
I am a software engineer from Japan, specializing in machine learning, natural language processing, and information retrieval.
I am also engaged in research on efficiency in a wide range of applications, such as data mining, geographic information systems, and natural language processing, by leveraging advanced data structures and string processing techniques. See Research for his achievements.
My handle name is Kampersanda, which is derived as: Kanda → K and A → K & A → K ampersand A → Kampersanda.
Technical interest
- String processing
- Compact data structures
- Code optimization
- Similarity search
- Information retrieval
- Natural language processing
- Machine learning
Career
- Aug. 2024 – present: Software engineer at Cierpa & Co., Inc.
- Oct. 2021 – Jul. 2024: Senior software/research engineer at LegalOn Technologies, Inc.
- Apr. 2018 – Sep. 2021: Postdoctoral researcher of Succinct Information Processing Unit at RIKEN AIP
- Apr. 2017 – Mar. 2018: JSPS research fellowship, DC2
- Apr. 2016 – Mar. 2018: PhD student in Graduate School of Advanced Technology and Science, Tokushima University (Early completion)
- Apr. 2014 – Mar. 2016: Master student in Graduate School of Advanced Technology and Science, Tokushima University
Recent activities
- 2024-11-09: Presented a talk, Leveraging LLMs for Unsupervised Dense Retriever Ranking (紹介), in IR Reading 2024秋
- 2024-06-26: Posted a preprint, Vaporetto: Efficient Japanese Tokenization Based on Improved Pointwise Linear Classification, on arXiv
- 2024-05-31: Presented a talk, Lucene/Elasticsearch の Character Filter でユニコード正規化するとトークンのオフセットがズレるバグへの Workaround, in Search Engineering Tech Talk 2024 Spring
- 2024-05-31: Posted an article, Lucene/Elasticsearch の Character Filter でユニコード正規化するとトークンのオフセットがズレるバグへの Workaround, on LegalOn Technologies Engineering Blog
- 2024-05-17: Posted a slide, Binary and Scalar Embedding Quantization for Significantly Faster & Cheaper Retrieval, on Speaker Deck
- 2024-03-25: Posted an article, Jaccard係数に基づく類似文書検索の高速化技法, on LegalOn Technologies Engineering Blog
- 2024-03-07: Posted a preprint, NP-Completeness for the Space-Optimality of Double-Array Tries, on arXiv
- 2023-12-09: Posted an article, SIF/uSIFを使ってRustで簡単高速文埋め込み, on my personal blog
- 2023-11-09: Posted an article, 社内勉強会で使用したSimCSEのチュートリアル資料を公開しました, on LegalOn Technologies Engineering Blog
Qualifications
- 統計検定2級(2024年8月)
- 応用情報技術者(2013年12月)
External links
Contact
- shnsk.knd (at) gmail.com
- Twitter DM (@kampersanda)