Thesis (Ph.D.)--Drexel University, 2010. / Includes abstract and vita. Includes bibliographical references (leaves 141-150).

Thesis (M.Phil.)--Hong Kong University of Science and Technology, 2005. / Includes bibliographical references (leaves 82-84). Also available in electronic version.

Thesis (Ph.D.)--Hong Kong University of Science and Technology, 2005. / Includes bibliographical references (leaves 100-106). Also available in electronic version.

Thesis (Ph. D.)--University of Texas at Austin, 2007. / Vita. Includes bibliographical references.

Thesis (M.S.)--University of Texas at Dallas, 2007. / Includes vita. Includes bibliographical references (leaves 81-84)

Thesis (Ph. D.)--University of Texas at Austin, 2006. / Vita. Includes bibliographical references.

Thesis (Ph.D.)--University of Texas at Dallas, 2007. / Includes vita. Includes bibliographical references (leaves 154-160)

In human-human dialogues, face-to-face meetings are often preferred over phone conversations.One explanation is that non-verbal modalities such as gesture provide additionalinformation, making communication more efficient and accurate. If so, computerprocessing of natural language could improve by attending to non-verbal modalitiesas well. We consider the problem of sentence segmentation, using hand-annotatedgesture features to improve recognition. We find that gesture features correlate wellwith sentence boundaries, but that these features improve the overall performance of alanguage-only system only marginally. This finding is in line with previous research onthis topic. We provide a regression analysis, revealing that for sentence boundarydetection, the gestural features are largely redundant with the language model andpause features. This suggests that gestural features can still be useful when speech recognition is inaccurate.

Social media users generate a large volume of data, which can contain meaningful and useful information. One such example is information about locations, which may be useful in applications such as marketing and security monitoring. There are two types of locations: location entities mentioned in the text of the messages and the physical locations of users. Extracting the first type of locations is not trivial because the location entities in the text are often ambiguous. In this thesis, we implement a sequential classification model with conditional random fields followed by a rule-based disambiguation model, we apply them to Twitter messages (tweets) and we show that they handle the ambiguous location entities in our dataset reasonably well. Only very few users disclose their physical locations; in order to automatically detect their locations, many approaches have been proposed using various types of information, including the tweets posted by the users. It is not easy to infer the original locations from text data, because text tends to be noisy, particularly in social media. Recently, deep learning techniques have been shown to reduce the error rate of many machine learning tasks, due to their ability to learn meaningful representations of input data. We investigate the potential of building a deep-learning architecture to infer the location of Twitter users based merely on their tweets. We find that stacked denoising auto-encoders are well suited for this task, with results comparable to state-of-the-art models. Finally, we combine the two models above with a third-party sentiment analysis tool and obtain a intelligent social media monitoring system. We show a demo of the system and that it is able to predict and visualize the locations and sentiments contained in a stream of tweets related to mobile phone brands - a typical real world e-business application.

Text classification is a wide research field with existing ready-to-use solutions for supervised training of text classifiers. The task of classifying short texts puts dif-ferent demands on the invoked learning system that general text classification does not. This thesis explores this challenge by experimenting on how to design the clas-sification system and what text features granted the best results. In the experimental study, a hierarchical versus a flat design was compared, along with different aspects of text features. The method consisted of training and testing on a dataset of 3.2 million samples in total. The test results were evaluated with the quality measures: precision, recall, F1-score and ROC analysis with a modification to target multi-class classification. The result of the experimental study was: 2-level hierarchical designed classifier gave better results than a flat designed classifier in 11 out of 13 occasions; integer represented terms outperformed TFIDF weighted terms of BOW features; lowercase conversion improved the classification results; bigram and tri-gram BOW features achieved better results than unigram BOW features. The results of the experimental study were used in a case study together with Thingmap, which maps natural language queries with users. The case study showed an improvement over earlier solutions of Thingmap’s system.