about

Data Science and Machine Intelligence Lab (DSMI Lab) locates in SA314 at National Chiao Tung University. Our researches focus on data science, machine learning, and deep learning. The following are our main research topics in recent years:

• Natural Language Processing:
  We focus on disinformation and Chatbot system for customer survice. The disinformation project aims to help reader determine whether the information is true or not. Recently, we focus on the research of clickbait analysis and title-body text similarity.

• Smart Green House
  We mainly cooperate with Taiwan Agricultural Research Institute Council of Agriculture. For the first-year project, we utilize the uniform experimental design to allocate the synthetic sensors. We estimate those synthetic sensor readings on the basis of linear model locally. We then apply $\epsilon$-SSVR to fit the globally three-dimensional heat map by combining real sensor and synthetic sensor readings.
  The following is the link of the paper:
  https://www.mdpi.com/2504-3900/31/1/63.
  For the second-year project, our ultimate goal is to induce the micro-weather of the greenhouse from outside in by the meteorological data forecast by Central Weather Bureau; before directly applying the meteorological data, our first step is to examine the feasibility of deriving each of inner readings (i.e., temperature, humidity or luminosity) purely from those of outer weather stations.

• Anomaly detection
  We propose an online oversampling principal component analysis (osPCA) algorithm to detecting outliers from a large amount of data via an online updating technique. Unlike prior principal component analysis (PCA)-based approaches, we do not store the entire data matrix or covariance matrix, and thus our approach is especially of interest in online or large-scale problems.
  The link of the published paper is as follows:
  https://ieeexplore.ieee.org/abstract/document/6200273

• Distributed Learning:
  As …, we applied distributed…. on SVM and PCA.

• SVM
  In DSMI lab, we are intrested in SVM, SSVM (Smooth Support Vector Machine), and RSVM (Reduced Support Vector Machine).

• Security:
  In the era of big data, it is a very important issue to assist information security with more effective legal detection and protection through the technology of data science. The current research is to classify malicious programs through deep learning technology. Through the deep learning technology can automatic feature extraction, it can quickly learn and identify different malicious program families, which can reduce the cost of past security experts in analyzing malicious programs, and we aim to assist security experts focus on the suspicious part of the malware.

• Time Series

Alumni

• 孫茂勛
• 黃彥傑
• 楊于萱
• 陳麒宇
• 翁玄懷

  Members

• 謝幸娟
• 吳芷綺
• 邱弋瑋
• 柯柏宏
• 李承穎
• 林永祥
• 邱信程
• 鄭翊騏
• 莊沅璟
• 張巧蓁

Advisor

• Data Science and Machine Intelligence Laboratory
• Yuh Jye Lee: yuhjye@math.nctu.edu.tw
• SA 314, No.1001, Daxue Rd., East Dist., Hsinchu City 300, Taiwan (R.O.C.)