2016学期视觉计算实验室第1周论文研读预告

时间:3月11日上午9:00

地点:基础教学楼B318实验室

研读成员朱秋辉,汪文涛。

一、朱秋辉
  交通工具的推断一直是一个很流行的话题,近些年来,越来越多的文章关注于各种交通方式固有的特点,本次介绍三篇文章:
  1.《urban sensing using smartphones for transportation mode classification》
  其思想是基于walk活动的路段来分割GPS轨迹来推断,创新点:
  (1)A key finding of our work is that walking activity can be robustly detected and acts as a separator for partitioning the data into single mode.
  (2)Our approach yields high accuracy despite the low sampling interval and does not require GPS data。
  (3)Device power consumption is effectively minimized。
  2.《Transportaion Mode Annotation of Tourist GPS trajectories Under environmental constraints》
  文章思想根据现今城市的设施(如train只适合在轨道上行驶)来进行推断。创新点:A new, simple GPS semantic annotation method using environmental constraints without machine learning. It is necessary for machine learning to label the data for training, the requirement is costly.
  3.《Comparison of advanced imputation algorithms for detection of transportation mode and activity episode using GPS data》
对于GPS数据进行推断交通工具时比较了各种分类方法,得出Bayesian network是最好的分类模型。

二、汪文涛
  1.《Influence Factor Based Opinion Mining of Twitter Data Using Supervised Learning》
  内容:This paper proposes a hybrid approach of extracting opinion using direct and indirect features of Twitter data based on Support Vector Machines (SVM), Naive Bayes,Maximum Entropy and Artificial Neural Networks based supervised classifiers.
  推荐理由:利用用户在Twitter的影响因素来预测选举结果,想法非常新颖。
  2.《Opinion Mining and Sentiment Analysis on a Twitter Data Stream》
  内容:This paper discusses an approach where a publicised stream of tweets from the Twitter microblogging site are preprocessed and classified based on their emotional content as positive, negative and irrelevant; and analyses the performance of various classifying algorithms based on their precision and recall in such cases.
  推荐理由:在不同大小数据集上对不同分类算法的进行了比较。
  3.《A Semi-supervised Fuzzy Co-clustering Framework and Application to Twitter Data Analysis》
  内容:In this paper, a novel framework for performing fuzzy co-clustering of cooccurrence information with partial supervision is proposed, which is induced by multinomial mixture concept.
  推荐理由:平时我们接触的要么是有监督学习,要么是无监督学习,想通过这篇文章对半监督学习加强理解。

123