Verb classifications have attracted a great deal of interest in both linguistics and natural language processing (NLP). They have proved useful for important tasks and applications, including e.g. computational lexicography, parsing, word sense disambiguation, semantic role labelling, information extraction, question-answering, and machine translation (Swier and Stevenson, 2004; Dang, 2004; Shi and Mihalcea, 2005; Kipper et al., 2008; Zapirain et al., 2008; Rios et al., 2011). Particularly useful are classes which capture generalizations about a range of linguistic properties (e.g. lexical, (morpho-)syntactic, semantic), such as those proposed by Beth Levin (1993). However, full exploitation of such classes in real-world tasks has been limited because no comprehensive or domain-specific lexical classification is available. This thesis investigates how Levin-style lexical semantic classes could be learned automatically from corpus data. Automatic acquisition is cost-effective when it involves either no or minimal supervision and it can be applied to any domain of interest where adequate corpus data is available. We improve on earlier work on automatic verb clustering. We introduce new features and new clustering methods to improve the accuracy and coverage. We evaluate our methods and features on well-established cross-domain datasets in English, on a specific domain of English (the biomedical) and on another language (French), reporting promising results. Finally, our task-based evaluation demonstrates that the automatically acquired lexical classes enable new approaches to some NLP tasks (e.g. metaphor identification) and help to improve the accuracy of existing ones (e.g. argumentative zoning).

Toxic compounds, such as pesticides, are routinely tested against a range of aquatic, avian and mammalian species as part of the registration process. The need for reducing dependence on animal testing has led to an increasing interest in alternative methods such as in silico modelling. The QSAR (Quantitative Structure Activity Relationship)-based models are already in use for predicting physicochemical properties, environmental fate, eco-toxicological effects, and specific biological endpoints for a wide range of chemicals. Data plays an important role in modelling QSARs and also in result analysis for toxicity testing processes. This research addresses number of issues in predictive toxicology. One issue is the problem of data quality. Although large amount of toxicity data is available from online sources, this data may contain some unreliable samples and may be defined as of low quality. Its presentation also might not be consistent throughout different sources and that makes the access, interpretation and comparison of the information difficult. To address this issue we started with detailed investigation and experimental work on DEMETRA data. The DEMETRA datasets have been produced by the EC-funded project DEMETRA. Based on the investigation, experiments and the results obtained, the author identified a number of data quality criteria in order to provide a solution for data evaluation in toxicology domain. An algorithm has also been proposed to assess data quality before modelling. Another issue considered in the thesis was the missing values in datasets for toxicology domain. Least Square Method for a paired dataset and Serial Correlation for single version dataset provided the solution for the problem in two different situations. A procedural algorithm using these two methods has been proposed in order to overcome the problem of missing values. Another issue we paid attention to in this thesis was modelling of multi-class data sets in which the severe imbalance class samples distribution exists. The imbalanced data affect the performance of classifiers during the classification process. We have shown that as long as we understand how class members are constructed in dimensional space in each cluster we can reform the distribution and provide more knowledge domain for the classifier.

新聞事件可描述為「一個時間區間內、同一主題的相似新聞之集合」，而新聞大多僅是一完整事件的零碎片段，其內容也易受到媒體立場或撰寫角度不同有所差異；除此之外，龐大的新聞量亦使得想要瞭解事件全貌的困難度大增。因此，本研究將利用文字探勘技術群聚相關新聞為事件，以增進新聞所帶來的價值。 分類分群為文字探勘中很常見的步驟，亦是本研究將新聞群聚成事件所運用到的主要方法。最近鄰 (k-nearest neighbor, kNN)搜尋法可視為分類法中最常見的演算法之一，但由於kNN在分類上必須要每篇新聞兩兩比較並排序才得以選出最近鄰，這也產生了kNN在實作上的效能瓶頸。本研究提出了一個「建立距離參考基準點」的方法RTD-based kNN (Relative Text-Distance-based kNN)，透過在向量空間中建立一個基準點，讓所有文件利用與基準點的相對距離建立起遠近的關係，使得在選取前k個最近鄰之前，直接以相對關係篩選出較可能的候選文件，進而選出前k個最近鄰，透過相對距離的概念減少比較次數以改善效率。 本研究於Google News中抽取62個事件(共742篇新聞)，並依其分群結果作為測試與評估依據，以比較RTD-based kNN與kNN新聞事件分群時的績效。實驗結果呈現出RTD-based kNN的基準點以常用字字彙建立較佳，分群後的再合併則有助於改善結果，而在RTD-based kNN與kNN的F-measure並無顯著差距(α=0.05)的情況下，RTD-based kNN的運算時間低於kNN達28.13%。顯示RTD-based kNN能提供新聞事件分群時一個更好的方法。最後，本研究提供一些未來研究之方向。 / News Events can be described as "the aggregation of many similar news that describe the particular incident within a specific timeframe". Most of news article portraits only a part of a passage, and many of the content are bias because of different media standpoint or different viewpoint of reporters; in addition, the massive news source increases complexity of the incident. Therefore, this research paper employs Text Mining Technique to cluster similar news to a events that can value added a news contributed. Classification and Clustering technique is a frequently used in Text Mining, and K-nearest neighbor(kNN) is one of most common algorithms apply in classification. However, kNN requires massive comparison on each individual article, and it becomes the performance bottlenecks of kNN. This research proposed Relative Text-Distance-based kNN(RTD-based kNN), the core concept of this method is establish a Base, a distance reference point, through a Vector Space, all documents can create the distance relationship through the relative distance between itself and base. Through the concept of relative distance, it can decrease the number of comparison and improve the efficiency. This research chooses a sample of 62 events (with total of 742 news articles) from Google News for the test and evaluation. Under the condition of RTD-based kNN and kNN with a no significant difference in F-measure (α=0.05), RTD-based kNN out perform kNN in time decreased by 28.13%. This confirms RTD-based kNN is a better method in clustering news event. At last, this research provides some of the research aspect for the future.

In the last decade, a lot of machine learning and data mining based approaches have been used in the areas of intrusion detection, malware detection and classification and also traffic analysis. In the area of malware analysis, static binary analysis techniques have become increasingly difficult with the code obfuscation methods and code packing employed when writing the malware. The behavior-based analysis techniques are being used in large malware analysis systems because of this reason. In prior art, a number of clustering and classification techniques have been used to classify the malwares into families and to also identify new malware families, from the behavior reports. In this thesis, we have analysed in detail about the use of Profile Hidden Markov models for the problem of malware classification and clustering. The advantage of building accurate models with limited examples is very helpful in early detection and modeling of malware families. The thesis also revisits the learning setting of an Intrusion Detection System that employs machine learning for identifying attacks and normal traffic. It substantiates the suitability of incremental learning setting(or stream based learning setting) for the problem of learning attack patterns in IDS, when large volume of data arrive in a stream. Related to the above problem, an elaborate survey of the IDS that use data mining and machine learning was done. Experimental evaluation and comparison show that in terms of speed and accuracy, the stream based algorithms perform very well as large volumes of data are presented for classification as attack or non-attack patterns. The possibilities for using stream algorithms in different problems in security is elucidated in conclusion.