:::

研討會總覽

【Big data workshop】Learning to Analyze Big Data: Some Personal Experiences (Chinese presentation)

2017/09/06
  • 研討會日期 : 2017-09-06
  • 時間 : 11:00
  • 主講人 : Professor Henry Jin-Lung Lin (林金龍)
  • 主持人 : Professor Yeutien Chou
  • 地點 : Conference Room B110
  • 演講者簡介 : Professor Lin received his Ph.D. in Economics from University of California, San Diego in 1991. He is currently a Professor of Finance and the Chairman of FinTech Big Data Research Center at National Dong Hwa University. His research interests include time series analysis, econometrics, high frequency financial data analysis and forecasting.
  • 演講摘要 : The term big data has become increasingly prevalent, and it appears that economists need to know something about it. Yet, as big data involves massive data collection, processing, cleaning, feature extraction, analysis and presentation, it demands deep knowledge about web crawling, processing tools and data analytics, not to mention appropriate statistical methods. As a beginner in this field, I have spent a great deal of efforts and time to learn about mastering big data. In this talk, I would like to pass on my learning experience. One can easily get drowned in the big data ocean as there are so many materials surfing the net, including news, blogs, slides, video clips, tutorials, codes, conference and journal papers. Thus, I shall give the talk in a question-answer format. Hopefully, by answering the frequently asked questions, I could draw a clear picture about big data and shed some lights on learning to crunch it. Here are the questions. 1. What is big data? Why is big data so hot? 2. What are the similarities and differences between data mining and statistical analysis? 3. How is big data different from small data? 4. What are data mining, machine learning, deep learning and artificial intelligence? 5. Is it possible to use machine learning to perform causal analysis? 6. How important is deep learning? 7. How useful are text mining and time series mining? 8. How to handle big (massive) data? Parallel computing? Distributed file system? 9. Choosing between R, Python or Java, or all of them? 10. Bigdata analytics in R? To my mind, the best way to learn about big data is to analyze one. I shall cover several real examples in this talk.