SEARCH

Search Details

XIANG Kun
Faculty of Science and Engineering　Department of Applied Informatics
Research Associate

Researcher information

■ Degree
  • Science and Engineering Ph.D, Hosei University, Sep. 2024
  • Science and Engineering Master, Hosei University, Sep. 2021
■ Research Keyword
  • DNN構造(軽量化、少量データ)
  • Knowledge Distillation
  • 観光情報、Smart Tourism
  • 機械学習、深層学習、自然言語処理

Career

■ Career
  • Nov. 2024 - Present
    Hosei University, 教務助手
  • 2021 - 2024
    Hosei University, 博士後期 修了(工学)
  • 2019 - 2021
    Hosei University, 博士前期 修了
  • 2015 - 2019
    北京林業大学
  • 2017 - 2018
    イギリスブライトン大学
■ Educational Background
  • Sep. 2021 - Sep. 2024, Hosei University, Science and Engineering Ph.D, Applied Informatics
  • 2019 - 2021, Hosei University, Faculty of Science and Engineering, Applied Informatics
  • 2015 - 2019, Beijing Forestry University, 理学観光学
  • 2017 - 2018, Brighton University England

Research activity information

■ Paper
  • Tourist Information Engagement Trial　　　　　　　　　　　　　　　
    2025
  • ASAUN-CS: Adaptive Stage Activated Unfolding Network for Compressive Sensing　　　　　　　　　　　　　　　
    02 Jul. 2024
  • HAMCap: A Weak-Supervised Hybrid Attention-Based Capsule Neural Network for Fine-Grained Climate Change Debate Analysis　　　　　　　　　　　　　　　
    Oct. 2023
  • DARE: Distill and Reinforce Ensemble Neural Networks for Climate-Domain Processing　　　　　　　　　　　　　　　
    Apr. 2023
  • A Novel Rigorous Measurement Model for Big Data Quality Characteristics　　　　　　　　　　　　　　　
    Dec. 2022
  • Economic Trend Prediction based on DNN with Few shot Dataset　　　　　　　　　　　　　　　
    25 Aug. 2022
  • Ensemble Compressed Language Model Based on Knowledge Distillation and Multi-Task Learning　　　　　　　　　　　　　　　
    08 Jun. 2022
  • Sentiment Classification Method Based on Blending of Emoticons and Short Texts　　　　　　　　　　　　　　　
    12 Mar. 2022
  • Reputation Analysis Based on Weakly-Supervised Bi-LSTM-Attention Network　　　　　　　　　　　　　　　
    04 Aug. 2021
■ Lectures, oral presentations, etc.
  • The Dilemma and Future of Large Language Models　　　　　　　　　　　　　　　
    22 Nov. 2023
  • Reputation Analysis Based on Weakly-Supervised Bi-LSTM-Attention Network　　　　　　　　　　　　　　　
    Kun XIANG
    The 1st Univ.Carthage-Hosei International Joint Webinar, 17 Mar. 2021