Relevance feature discovery for text analysis

It is a big challenge to guarantee the quality of discovered relevance features in text documents for describing user preferences because of the large number of terms, patterns, and noise. Most existing popular text mining and classification methods have adopted term-based approaches. However, they have all suffered from the problems of polysemy and synonymy. Over the years, people have often held the hypothesis that pattern-based methods should perform better than term- based ones in describing user preferences, but many experiments do not support this hypothesis. This research presents a promising method, Relevance Feature Discovery (RFD), for solving this challenging issue. It discovers both positive and negative patterns in text documents as high-level features in order to accurately weight low-level features (terms) based on their specificity and their distributions in the high-level features. The thesis also introduces an adaptive model (called ARFD) to enhance the exibility of using RFD in adaptive environment. ARFD automatically updates the system's knowledge based on a sliding window over new incoming feedback documents. It can efficiently decide which incoming documents can bring in new knowledge into the system. Substantial experiments using the proposed models on Reuters Corpus Volume 1 and TREC topics show that the proposed models significantly outperform both the state-of-the-art term-based methods underpinned by Okapi BM25, Rocchio or Support Vector Machine and other pattern-based methods.

Scoring-thresholding pattern based text classifier

A big challenge for classification on text is the noisy of text data. It makes classification quality low. Many classification process can be divided into two sequential steps scoring and threshold setting (thresholding). Therefore to deal with noisy data problem, it is important to describe positive feature effectively scoring and to set a suitable threshold. Most existing text classifiers do not concentrate on these two jobs. In this paper, we propose a novel text classifier with pattern-based scoring that describe positive feature effectively, followed by threshold setting. The thresholding is based on score of training set, make it is simple to implement in other scoring methods. Experiment shows that our pattern-based classifier is promising.

Mapping semantic knowledge for unsupervised text categorisation

Text categorisation is challenging, due to the complex structure with heterogeneous, changing topics in documents. The performance of text categorisation relies on the quality of samples, effectiveness of document features, and the topic coverage of categories, depending on the employing strategies; supervised or unsupervised; single labelled or multi-labelled. Attempting to deal with these reliability issues in text categorisation, we propose an unsupervised multi-labelled text categorisation approach that maps the local knowledge in documents to global knowledge in a world ontology to optimise categorisation result. The conceptual framework of the approach consists of three modules; pattern mining for feature extraction; feature-subject mapping for categorisation; concept generalisation for optimised categorisation. The approach has been promisingly evaluated by compared with typical text categorisation methods, based on the ground truth encoded by human experts.

A pattern based two-stage text classifier

In a classification problem typically we face two challenging issues, the diverse characteristic of negative documents and sometimes a lot of negative documents that are closed to positive documents. Therefore, it is hard for a single classifier to clearly classify incoming documents into classes. This paper proposes a novel gradual problem solving to create a two-stage classifier. The first stage identifies reliable negatives (negative documents with weak positive characteristics). It concentrates on minimizing the number of false negative documents (recall-oriented). We use Rocchio, an existing recall based classifier, for this stage. The second stage is a precision-oriented “fine tuning”, concentrates on minimizing the number of false positive documents by applying pattern (a statistical phrase) mining techniques. In this stage a pattern-based scoring is followed by threshold setting (thresholding). Experiment shows that our statistical phrase based two-stage classifier is promising.

The heterogeneous cluster ensemble method using hubness for clustering text documents

We propose a cluster ensemble method to map the corpus documents into the semantic space embedded in Wikipedia and group them using multiple types of feature space. A heterogeneous cluster ensemble is constructed with multiple types of relations i.e. document-term, document-concept and document-category. A final clustering solution is obtained by exploiting associations between document pairs and hubness of the documents. Empirical analysis with various real data sets reveals that the proposed meth-od outperforms state-of-the-art text clustering approaches.

Rough set based approach to text classification

Textual document set has become an important and rapidly growing information source in the web. Text classification is one of the crucial technologies for information organisation and management. Text classification has become more and more important and attracted wide attention of researchers from different research fields. In this paper, many feature selection methods, the implement algorithms and applications of text classification are introduced firstly. However, because there are much noise in the knowledge extracted by current data-mining techniques for text classification, it leads to much uncertainty in the process of text classification which is produced from both the knowledge extraction and knowledge usage, therefore, more innovative techniques and methods are needed to improve the performance of text classification. It has been a critical step with great challenge to further improve the process of knowledge extraction and effectively utilization of the extracted knowledge. Rough Set decision making approach is proposed to use Rough Set decision techniques to more precisely classify the textual documents which are difficult to separate by the classic text classification methods. The purpose of this paper is to give an overview of existing text classification technologies, to demonstrate the Rough Set concepts and the decision making approach based on Rough Set theory for building more reliable and effective text classification framework with higher precision, to set up an innovative evaluation metric named CEI which is very effective for the performance assessment of the similar research, and to propose a promising research direction for addressing the challenging problems in text classification, text mining and other relative fields.

Injury narrative text classification : a preliminary study

Description of a patient's injuries is recorded in narrative text form by hospital emergency departments. For statistical reporting, this text data needs to be mapped to pre-defined codes. Existing research in this field uses the Naïve Bayes probabilistic method to build classifiers for mapping. In this paper, we focus on providing guidance on the selection of a classification method. We build a number of classifiers belonging to different classification families such as decision tree, probabilistic, neural networks, and instance-based, ensemble-based and kernel-based linear classifiers. An extensive pre-processing is carried out to ensure the quality of data and, in hence, the quality classification outcome. The records with a null entry in injury description are removed. The misspelling correction process is carried out by finding and replacing the misspelt word with a soundlike word. Meaningful phrases have been identified and kept, instead of removing the part of phrase as a stop word. The abbreviations appearing in many forms of entry are manually identified and only one form of abbreviations is used. Clustering is utilised to discriminate between non-frequent and frequent terms. This process reduced the number of text features dramatically from about 28,000 to 5000. The medical narrative text injury dataset, under consideration, is composed of many short documents. The data can be characterized as high-dimensional and sparse, i.e., few features are irrelevant but features are correlated with one another. Therefore, Matrix factorization techniques such as Singular Value Decomposition (SVD) and Non Negative Matrix Factorization (NNMF) have been used to map the processed feature space to a lower-dimensional feature space. Classifiers with these reduced feature space have been built. In experiments, a set of tests are conducted to reflect which classification method is best for the medical text classification. The Non Negative Matrix Factorization with Support Vector Machine method can achieve 93% precision which is higher than all the tested traditional classifiers. We also found that TF/IDF weighting which works well for long text classification is inferior to binary weighting in short document classification. Another finding is that the Top-n terms should be removed in consultation with medical experts, as it affects the classification performance.

Decision boundary setting and classifier combination for text classification

This thesis presents a promising boundary setting method for solving challenging issues in text classification to produce an effective text classifier. A classifier must identify boundary between classes optimally. However, after the features are selected, the boundary is still unclear with regard to mixed positive and negative documents. A classifier combination method to boost effectiveness of the classification model is also presented. The experiments carried out in the study demonstrate that the proposed classifier is promising.

Centroid training to achieve effective text classification

Traditional text classification technology based on machine learning and data mining techniques has made a big progress. However, it is still a big problem on how to draw an exact decision boundary between relevant and irrelevant objects in binary classification due to much uncertainty produced in the process of the traditional algorithms. The proposed model CTTC (Centroid Training for Text Classification) aims to build an uncertainty boundary to absorb as many indeterminate objects as possible so as to elevate the certainty of the relevant and irrelevant groups through the centroid clustering and training process. The clustering starts from the two training subsets labelled as relevant or irrelevant respectively to create two principal centroid vectors by which all the training samples are further separated into three groups: POS, NEG and BND, with all the indeterminate objects absorbed into the uncertain decision boundary BND. Two pairs of centroid vectors are proposed to be trained and optimized through the subsequent iterative multi-learning process, all of which are proposed to collaboratively help predict the polarities of the incoming objects thereafter. For the assessment of the proposed model, F1 and Accuracy have been chosen as the key evaluation measures. We stress the F1 measure because it can display the overall performance improvement of the final classifier better than Accuracy. A large number of experiments have been completed using the proposed model on the Reuters Corpus Volume 1 (RCV1) which is important standard dataset in the field. The experiment results show that the proposed model has significantly improved the binary text classification performance in both F1 and Accuracy compared with three other influential baseline models.

Query Click and Text Similarity Graph for Query Suggestions

Query suggestion is an important feature of the search engine with the explosive and diverse growth of web contents. Different kind of suggestions like query, image, movies, music and book etc. are used every day. Various types of data sources are used for the suggestions. If we model the data into various kinds of graphs then we can build a general method for any suggestions. In this paper, we have proposed a general method for query suggestion by combining two graphs: (1) query click graph which captures the relationship between queries frequently clicked on common URLs and (2) query text similarity graph which finds the similarity between two queries using Jaccard similarity. The proposed method provides literally as well as semantically relevant queries for users' need. Simulation results show that the proposed algorithm outperforms heat diffusion method by providing more number of relevant queries. It can be used for recommendation tasks like query, image, and product suggestion.

Technical Vocabulary and Medieval Text Types: A Semantic Field Approach

This paper seeks to discover in what sense we can classify vocabulary items as technical terms in the later medieval period. In order to arrive at a principled categorization of technicality, distribution is taken as a diagnostic factor: vocabulary shared across the widest range of text types may be assumed to be both prototypical for the semantic field, but also the most general and therefore least technical terms since lexical items derive at least part of their meaning from context, a wider range of contexts implying a wider range of senses. A further way of addressing the question of technicality is tested through the classification of the lexis into semantic hierarchies: in the terms of componential analysis, having more components of meaning puts a term lower in the semantic hierarchy and flags it as having a greater specificity of sense, and thus as more technical. The various text types are interrogated through comparison of the number of levels in their hierarchies and number of lexical items at each level within the hierarchies. Focusing on the vocabulary of a single semantic field, DRESS AND TEXTILES, this paper investigates how four medieval text types (wills, sumptuary laws, petitions, and romances) employ technical terminology in the establishment of the conventions of their genres.

Improvements on the Porter`s Stemming Algorithm for Portuguese

The amount of textual information digitally stored is growing every day. However, our capability of processing and analyzing that information is not growing at the same pace. To overcome this limitation, it is important to develop semiautomatic processes to extract relevant knowledge from textual information, such as the text mining process. One of the main and most expensive stages of the text mining process is the text pre-processing stage, where the unstructured text should be transformed to structured format such as an attribute-value table. The stemming process, i.e. linguistics normalization, is usually used to find the attributes of this table. However, the stemming process is strongly dependent on the language in which the original textual information is given. Furthermore, for most languages, the stemming algorithms proposed in the literature are computationally expensive. In this work, several improvements of the well know Porter stemming algorithm for the Portuguese language, which explore the characteristics of this language, are proposed. Experimental results show that the proposed algorithm executes in far less time without affecting the quality of the generated stems.

Seeing beyond reading: a survey on visual text analytics

We review recent visualization techniques aimed at supporting tasks that require the analysis of text documents, from approaches targeted at visually summarizing the relevant content of a single document to those aimed at assisting exploratory investigation of whole collections of documents.Techniques are organized considering their target input materialeither single texts or collections of textsand their focus, which may be at displaying content, emphasizing relevant relationships, highlighting the temporal evolution of a document or collection, or helping users to handle results from a query posed to a search engine.We describe the approaches adopted by distinct techniques and briefly review the strategies they employ to obtain meaningful text models, discuss how they extract the information required to produce representative visualizations, the tasks they intend to support and the interaction issues involved, and strengths and limitations. Finally, we show a summary of techniques, highlighting their goals and distinguishing characteristics. We also briefly discuss some open problems and research directions in the fields of visual text mining and text analytics.

Weakly supervised joint sentiment-topic detection from text

Sentiment analysis or opinion mining aims to use automated tools to detect subjective information such as opinions, attitudes, and feelings expressed in text. This paper proposes a novel probabilistic modeling framework called joint sentiment-topic (JST) model based on latent Dirichlet allocation (LDA), which detects sentiment and topic simultaneously from text. A reparameterized version of the JST model called Reverse-JST, obtained by reversing the sequence of sentiment and topic generation in the modeling process, is also studied. Although JST is equivalent to Reverse-JST without a hierarchical prior, extensive experiments show that when sentiment priors are added, JST performs consistently better than Reverse-JST. Besides, unlike supervised approaches to sentiment classification which often fail to produce satisfactory performance when shifting to other domains, the weakly supervised nature of JST makes it highly portable to other domains. This is verified by the experimental results on data sets from five different domains where the JST model even outperforms existing semi-supervised approaches in some of the data sets despite using no labeled documents. Moreover, the topics and topic sentiment detected by JST are indeed coherent and informative. We hypothesize that the JST model can readily meet the demand of large-scale sentiment analysis from the web in an open-ended fashion.

Text Analytics of Social Media: Sentiment Analysis, Event Detection and Summarization

In the last decade, large numbers of social media services have emerged and been widely used in people's daily life as important information sharing and acquisition tools. With a substantial amount of user-contributed text data on social media, it becomes a necessity to develop methods and tools for text analysis for this emerging data, in order to better utilize it to deliver meaningful information to users. Previous work on text analytics in last several decades is mainly focused on traditional types of text like emails, news and academic literatures, and several critical issues to text data on social media have not been well explored: 1) how to detect sentiment from text on social media; 2) how to make use of social media's real-time nature; 3) how to address information overload for flexible information needs. In this dissertation, we focus on these three problems. First, to detect sentiment of text on social media, we propose a non-negative matrix tri-factorization (tri-NMF) based dual active supervision method to minimize human labeling efforts for the new type of data. Second, to make use of social media's real-time nature, we propose approaches to detect events from text streams on social media. Third, to address information overload for flexible information needs, we propose two summarization framework, dominating set based summarization framework and learning-to-rank based summarization framework. The dominating set based summarization framework can be applied for different types of summarization problems, while the learning-to-rank based summarization framework helps utilize the existing training data to guild the new summarization tasks. In addition, we integrate these techneques in an application study of event summarization for sports games as an example of how to better utilize social media data.