Shou-De Lin

Chief Machine Learning Scientist, Appier

Professor
Department of Computer Science and Information Engineering
Graduate Institute of Networking and Multimedia
National Taiwan University

Tel: +886-2-33664888 ext.333, Fax: +886-2-33664898, Office: Rm333
Email:

Recruiting: We Train Big Data Scientists !!

Do you want to learn how to deal with Tera bytes of data coming from different sources?

Do you want to use data to predict what will happen in the future?

Do you hope to design analytic models that can be used in real life?

Do you feel excited about this area after reading my popular science article about big data?
Do you want to know how we can beat hundreds of teams in KDD Cup year by year?

If any of the above answer is yes, feel free to email me about the possibility to join MSLAB.

NEWS:
  • Slides: Research and Writing Tips for Graduate Students
  • MSLAB Summer 2016 Training
  • The paper “Inferring Air Quality for Station Location Recommendation Based on Urban Big Data” in KDD2015
  • The paper "A Learning-based Framework to handle Multi-round Multi-party influence maximization on social networks " in KDD2015
  • The paper "A Dual-Augmented Block Minimization Framework for Learning with Limited Memory " in NIPS2015
  • The paper "LambdaMF: Learning Nonsmooth Ranking Functions in Matrix Factorization Using Lambda" in ICDM2015
  • The paper "Matching Users and Items Across Domains to Improve the Recommendation Quality"  in KDD2014
  • The paper "Enriching Cold Start Personalized Language Model Using Social Network Information" in ACL2014
  • The paper "Robust Inverse Covariance Estimation under Noisy Measurements" accepted in ICML2014
  • The paper "Sparse Random Feature Algorithm as Coordinate Descent in Hilbert Space " accepted in NIPS2014
  • Congratulate Dr. Cheng-Te Li for becoming a research assistant fellow in Sinica !!
  • Congratulate Dr. Cheng-Te Li has received the best dissertation award in ACLCLP!!
  • A long Paper (11%) and a short paper (19%) Accepted in ICDM 2013
    • "Sampling Heterogeneous Networks"
    • “Communication-Efficient Distributed Multiple Reference Pattern Matching for M2M Systems”
  • Congratulate the NTU team wins KDD Cup 2013 Champions for both tracks. Our team (composed by different students every year) has won 5 KDD-Cup Champions from 2008~2013 except 2009 (3rd place).

kdd cup 2013kdd cup 2013

  • How to Write a proper SOP
  • Two regular papers accepted in KDD13 (17% acceptance rate), which reflects the hard working and smartness of our students:
        ”Indexed Block Coordinate Descent for Large-Scale Linear Classification with Limited Memory ”
        "Unsupervised Link Prediction Using Aggregative Statistics on Heterogeneous Social Networks”
  • Congratulate Dr. Hung-Yi Lo has become the assistant professor in Shih Chien University.
  • Congratulate the NTU team wins KDD Cup 2012 Track 2 Champion, this is the 5th consecutive win of our team from 2008-2012 (4 of them are champions).
  • News:
    • 台大校刊
    • INTEL NEWS
    • 本實驗室目前研究計畫:
      • 社群網路探勘
      • 自然語言處理
      • 機器學習與資料探勘
      • 感測網路資料分析
    • 本實驗室2012年新的研究計畫:
      • data mining and machine learning for security (利用資料探勘與機器學習方法進行資訊安全研究)
      • Location-based intelligent service (以地理位置為基礎的智慧型服務)
  • Champions for Both Tracks in KDDCUP 2011

          KDD Cup 2011 Plaque

For students and Postdoc/RA candidates

  • Please email me if you have interests to work on a project in our Lab. I will also be supervising new undergraduate and graduate students. You are more than welcome to contact me if you feel there is a potential match between our research interests.

Prospective projects

 

社群網路探勘 (SNA):

  • Electronic Holmes (電子福爾摩斯): Identifying suspicious instances from complex networks
  • Social Network Analysis (社群網路分析)
  • Plurk, Facebook, data analysis (PTT/FB 資料分析)
  • Social Voting Theory and Analysis (社群投票行為模型)

知識發現與機器學習(ML/KDD)

  • Knowledge Discovery and Data Mining (資料探勘相關) Time sequence and event detection, Imbalance Data Mining, KDDCUP, etc.
  • Machine Learning (機器學習): CRF, EM, Multi-label Learning, other applications
  • Machine Discovery (機器發明): Discover interesting things automatically
  • Network Forensics, Security with Machine learning

自然語言處理搜尋引擎,AI (NLP/IR)

  • AutoEdit (自動校定系統): Edit a flawed English sentence into a more accurate one
  • Connection Search Engine (連結搜索引擎): Finds social relationships from texts
  • Poetry Writing (古典詩詞創作)
  • Plagiarism Detection (抄襲偵測)
  • CS methods for popular science writing (資訊方法在寫作上的運用)

visitors