PEOPLE/

JapaneseJA
EnglishEN

Makoto Nakatsuji

Makoto Nakatsuji is a manager of the NTT Resonant, Inc. He completed his Ph.D. in Social Informatics at Kyoto University Graduate School of Informatics in 2010. He was a Visiting Schlor in The Tetherless World Constellation at Rensselaer Polytechnic Institute in 2013. His research interests include Question-answering systems, Deep learning algorithms, semantic data mininig, recommendation, and link prediction. He has applied the above techniques to the actual AI services such as love advice service by AI oshiel as well as AI chat bot service with drama main character.

Publication

  • Makoto Nakatsuji, Sohei Okui: Answer Generation through Unified Memories over Multiple Passages, IJCAI, (2020) paper (pdf)
  • Makoto Nakatsuji, Sohei Okui: Conclusion-Supplement Answer Generation for Non-Factoid Questions, AAAI, (2020)paper (pdf) 「恋愛相談オシエル」や「恋するAI歌人」、「AI菜奈ちゃん」の関連技術
  • Makoto Nakatsuji: Can AI Generate Love Advice?: Toward Neural Answer Generation for Non-Factoid Questions, arXiv:1912.10163 [cs.CL] (2019)paper (pdf)
  • Makoto Nakatsuji: Can AI Generate Love Advice? Neural Conclusion-Supplement Answer Generation for Non Factoid Questions, In GPU Technology Conference 2018, San Jose, CA, United states, (2018):slide site
  • Makoto Nakatsuji, QingPeng Zhang, Xio Lu, Bassem Makni, Jim Hendler: Semantic Social Network Analysis by Cross-Domain Tensor Factorization, IEEE Transactions on Computational Social Systems: vol. PP, no. 99, 1-11 (2017)paper (pdf)
  • Yasuhiro Fujiwara, Makoto Nakatsuji, Hiroaki Shiokawa, Takeshi Mishima, Makoto Onizuka: Fast Ad-Hoc Search Algorithm for Personalized PageRank, IEICE Transactions 100-D(4): 610-620 (2017)
  • Makoto Nakatsuji:Semantic Sensitive Simultaneous Tensor Factorization, ISWC (1) 2016: 411-427 (2016)
  • Makoto Nakatsuji, Hiroyuki Toda, Hiroshi Sawada, Jin Zheng, Jim Hendler: Semantic Sensitive Tensor Factorization, Artificial Intelligence Journal, Elsevier, 230: 224-245, Elsevier (2016) paper (pdf)code
  • Yasuhiro Fujiwara, Makoto Nakatsuji, Hiroaki Shiokawa, Yasutoshi Ida, Machiko Toyoda: Adaptive Message Update for Fast Affinity Propagation, KDD: 309-318 (2015)
  • Makoto Nakatsuji, Yasuhiro Fujiwara, Hiroyuki Toda, Hiroshi Sawada, Jin Zheng, Jim Hendler: Semantic Data Representation for Improving Tensor Factorization, AAAI: 2004-2012 (2014)
  • Makoto Nakatsuji, Yasuhiro Fujiwara: Linked Taxonomies to Capture Users' Subjective Assessments of Items to Facilitate Accurate Collaborative Filtering, Artificial Intelligence Journal, Elsevier, 207(0): 52-68, Elsevier (2014) paper (pdf)
  • Yasuhiro Fujiwara, Makoto Nakatsuji, Hiroaki Shiokawa, Takeshi Mishima, Makoto Onizuka: Fast and Exact Top-k Algorithm for PageRank, AAAI: 1106-1112 (2013)
  • Yasuhiro Fujiwara, Makoto Nakatsuji, Hiroaki Shiokawa, Takeshi Mishima, Makoto Onizuka: Efficient Ad-hoc Search for Personalized PageRank, SIGMOD: 445-456 (2013)
  • Yasuhiro Fujiwara, Makoto Nakatsuji, Hiroaki Shiokawa, Makoto Onizuka: Efficient Search Algorithm for SimRank, ICDE: 589-600 (2013)
  • Makoto Nakatsuji, Yasuhiro Fujiwara, Toshio Uchiyama, Hiroyuki Toda: Collaborative Filtering by Analyzing Dynamic User Interests Modeled by Taxonomy. ISWC 2012: 361-377 (2012) paper (pdf)
  • Makoto Nakatsuji, Akimichi Tanaka, Toshio Uchiyama, Ko Fujimura: Extracting Communities of Interests for Semantics-Based Graph Searches. IEICE Transactions 95-D(4): 932-941 (2012)
  • Yasuhiro Fujiwara, Makoto Nakatsuji, Makoto Onizuka, Masaru Kitsuregawa: Fast and Exact Top-k Search for Random Walk with Restart. PVLDB 5(5): 442-453 (2012)
  • Yasuhiro Fujiwara, Makoto Nakatsuji, Takeshi Yamamuro, Hiroaki Shiokawa, Makoto Onizuka: Efficient personalized pagerank with accuracy assurance. KDD 2012: 15-23
  • Makoto Nakatsuji, Yasuhiro Fujiwara, Toshio Uchiyama, Ko Fujimura: User Similarity from Linked Taxonomies: Subjective Assessments of Items. IJCAI 2011: 2305-2311 paper (pdf)
  • Makoto Nakatsuji. Identifying Novel Topics based on User Interests. In Atill Elci, Mamdou Koné, and Mehmet Orgun, editors, Semantic Agent Systems: Foundations and Applications, Studies in Computational Intelligence. Springer US, 273-292 (2011).
  • Makoto Nakatsuji, Akimichi Tanaka, Takahiro Madokoro, Kenichiro Okamoto, Sumio Miyazaki, Tadasu Uchiyama: Extracting Know-Who/Know-How Using Development Project-Related Taxonomies. IEICE Transactions 93-D(10): 2717-2727 (2010)
  • Makoto Nakatsuji, Yasuhiro Fujiwara, Akimichi Tanaka, Toshio Uchiyama, Ko Fujimura, Toru Ishida: Classical music for rock fans?: novel recommendations for expanding user interests. CIKM 2010: 949-958 paper (pdf)
  • Makoto Nakatsuji, Yasuhiro Fujiwara, Akimichi Tanaka, Tadasu Uchiyama, Toru Ishida: Recommendations Over Domain Specific User Graphs. ECAI 2010: 607-612 paper (pdf)
  • Kyota Tsutsumida, Jun Okamoto, Shun Ishizaki, Makoto Nakatsuji, Akimichi Tanaka, Tadasu Uchiyama: Study of Word Sense Disambiguation System that uses Contextual Features - Approach of Combining Associative Concept Dictionary and Corpus -. LREC 2010
  • Makoto Nakatsuji, Makoto Yoshida, Toru Ishida: Detecting innovative topics based on user-interest ontology. Journal of Web Semantics, 7(2): 107-120, Elsevier (2009)
  • Takahiro Madokoro, Makoto Nakatsuji, Kenichiro Okamoto, Sumio Miyazaki, Tsuyoshi Harada: Know-who/know-how Navigation Using Development Project-related Taxonomies. ICSC 2009: 559-560
  • Makoto Nakatsuji, Akimichi Tanaka, Toru Ishida: Extracting Communities of Interests for Semantics-based Graph Searches. International Semantic Web Conference (Posters & Demos) 2008
  • Makoto Nakatsuji, Yu Miyoshi, Yoshihiro Otsuka, Miki Hirano: Flexible Interface Mapping for System Cooperation and Its Evaluation. Transaction of Information Processing Society of Japan 48 (SIG_14(TOD_35)), 47-59, (2007) paper (pdf)
  • Makoto Nakatsuji, Minoru Kawahara, Hiroyuki Kawano: The architecture and performance of topic-driven P2P resource discovery system. Systems and Computers in Japan 38(5): 69-79 (2007)
  • Makoto Nakatsuji, Yu Miyoshi, Yoshihiro Otsuka: Innovation Detection Based on User-Interest Ontology of Blog Community. ISWC 2006: 515-528 (2006) paper (pdf)
  • Makoto Nakatsuji, Yu Miyoshi, Tatsuyuki Kimura: Proposal and Verification of Flexible Interface Mapping Technique for Automatic System Cooperation Based on Semantics. Web Intelligence 2005: 812-813
  • Makoto Nakatsuji, Minoru Kawahara, and Hiroyuki Kawano: Advanced index refinement by classifiers and distillers in P2P resource discovery, International Conference on Intelligent Agents Web Technology and Internet Commerce 2003, pp.272-285, Vienna, Austria, Feb. 2003.

Japanese Journal Publication

  • 中辻 真, 複数パッセージの関連性を考慮した回答生成手法, 人工知能学会論文誌: No. LC4, 第37巻4号, 2022.
  • 中辻 真，セマンティクスを用いた複数テンソルの同時分解手法, 電子情報通信学会論文誌：Vol.J105-D,No.06,pp.-,Jun. 2022.
  • 中辻 真, 八島 浩之, Non-Factoid型質問のための結論と理由で構成される回答文の生成手法, 人工知能学会論文誌: No. L64, 第37巻2号, 2021.
  • 中辻 真，奥井颯平，藤田明久，”LSTMを用いたNon-Factoid型長文回答構築手法”, 電子情報通信学会論文誌：Vol.J102-D,No.4,pp.-,Apr. 2019年 paper (pdf)
  • 中辻 真, ディープラーニング活用事例と使いこなしの勘所：［言語処理分野］4．AIによる恋愛相談への回答生成 -答えのない回答生成への試み- . 2018年
  • 乙守 信行, 中辻 真, 萩野 達也, "オープンデータの普及促進を加速させるコンテストの開催 ─ LOD チャレンジJapan の取組み─", 人工知能学会誌 特集：「Linked Data とセマンティック技術」 paper (pdf), Vol. 30, No .5, pp.591-598, 2015年9月
  • 片岡泰之，中辻 真，戸田浩之，小池義昌，松尾豊: "LOD とソーシャルメディアに基づく 特定行動に関するオントロジーの構築と評価", 人工知能学会論文誌, 2015, to appear
  • 中辻 真，藤原靖宏，戸田浩之，澤田 宏，チェン ジン，ヘンドラー ジェームズ: "セマンティクスを用いたテンソル分解手法", 人工知能学会論文誌,Vol. 30, No. 3, 2015年 paper (pdf)
  • 藤原 靖宏，中辻 真，塩川 浩昭，三島 健，鬼塚 真，"Personalized PageRank に対するアドホックな検索手法"，電子情報通信学会論文誌. 2015年
  • 藤原 靖宏, 中辻 真, 塩川 浩昭, 三島 健, 鬼塚 真: "PageRank のための高速なTop-k 検索", 人工知能学会論文誌,Vol. 30, No. 2, 2015年2月
  • 中辻 真, 藤原 靖宏, 内山 俊郎, 戸田 浩之: "動的なユーザ興味に対応したセマンティクスに基づく情報推薦手法", 人工知能学会論文誌,Vol. 28, No. 6, 2013年10月 paper (pdf)
  • 近藤 光正, 中辻真, 田中 明通: "Wikipediaに基づくWeb閲覧履歴からの潜在的興味キーワード抽出 Wikipedia-Based Latent Interest Keywords Extraction from Web Browsing History", (データマイニング,<特集>データ工学と情報マネジメント論文), 電子情報通信学会論文誌. D, 情報・システム J96-D(5), 1199-1211, 2013-05-01
  • 中辻 真, 藤原 靖宏, 内山 俊郎: "タクソノミを用いたNoveltyの高いアイテムの推薦手法", 電子情報通信学会論文誌 D, Vol.J96-D, No.4, pp.926-940, 2013年4月
  • 中辻 真, 藤原 靖宏, 内山 俊郎: "ユーザグラフ上のランダムウォークに基づくクロスドメイン推薦", 人工知能学会論文誌, Vol. 27, No .5, 2012年10月 paper (pdf)
  • 藤原 靖宏, 中辻 真, 鬼塚 真, 喜連川 優: "Random walk with restart に対する高速な検索手法", 情報処理学会論文誌:データベース, Vol. 4, No. 2, 2011年6月
  • 山登庸次, 中辻 真, 須永宏: "ユビキタス環境にて動的にサービス実現するためのサービス合成技術", 情報処理学会論文誌, 48(2), 562-577, 2007年2月
  • 中辻 真 川原 稔, 河野 浩之, "トピック主導型P2P情報検索システムの提案と性能評価", 電子情報通信学会論文誌. D-I, 情報・システム, I-情報処理 J87-D-I(2), 126-136, 2004年2月

Oral presentation

  • Makoto Nakatsuji: Can AI Generate Love Advice? Neural Conclusion-Supplement Answer Generation for Non Factoid Questions, In GPU Technology Conference 2018, San Jose, CA, United states, (2018): presentation
  • ［招待講演］キャラクターAIとの対話実現に向けた取組み, 第168回データベースシステム研究発表会, (2018)
  • 「教えて!goo」3000万件のQAデータから、世界初の長文生成AIが生まれるまで～AIによる恋愛相談の裏側～, Developers Summit 2018 Summer, 2018.7(2018)
  • ディープラーニングによる新たな対話AIサービス実践について (2017)

Media

  • Dear Oshieru: AI Tips on Love, NHK WORLD (2016)
  • 人間ってナンだ？超ＡＩ入門　第７回「恋愛する」
  • 人間ってナンだ？超ＡＩ入門　シーズン2　第7回「恋愛する」

Service involved