Atualmente, o meio mais comum de busca de informações é a WEB. Assim, é importante procurar métodos eficientes para recuperar essa informação. As máquinas de busca na WEB usualmente utilizam palavras-chaves para expressar uma busca. Porém, não é trivial caracterizar a informação desejada. Usuários diferentes com necessidades diferentes podem estar interessados em informações relacionadas, mas distintas, ao realizar a mesma busca. O processo de realimentação de relevância torna possível a participação ativa do usuário no processo de busca. A idéia geral desse processo consiste em, após o usuário realizar uma busca na WEB permitir que indique, dentre os sites encontrados, quais deles considera relevantes e não relevantes. A opinião do usuário pode então ser considerada para reordenar os dados, de forma que os sites relevantes para o usuário sejam retornados mais facilmente. Nesse contexto, e considerando que, na grande maioria dos casos, uma consulta retorna um número muito grande de sites WEB que a satisfazem, das quais o usuário é responsável por indicar um pequeno número de sites relevantes e não relevantes, tem-se o cenário ideal para utilizar aprendizado parcialmente supervisionado, pois essa classe de algoritmos de aprendizado requer um número pequeno de exemplos rotulados e um grande número de exemplos não-rotulados. Assim, partindo da hipótese que a utilização de aprendizado parcialmente supervisionado é apropriada para induzir um classificador que pode ser utilizado como um filtro de realimentação de relevância para buscas na WEB, o objetivo deste trabalho consiste em explorar algoritmos de aprendizado parcialmente supervisionado, mais especificamente, aqueles que utilizam multidescrição de dados, para auxiliar na recuperação de sites na WEB. Para avaliar esta hipótese foi projetada e desenvolvida uma ferramenta denominada C-SEARCH que realiza esta reordenação dos sites a partir da indicação do usuário. Experimentos mostram que, em casos que buscas genéricas, que o resultado possui um bom diferencial entre sites relevantes e irrelevantes, o sistema consegue obter melhores resultados para o usuário / As nowadays the WEB is the most common source of information, it is very important to find reliable and efficient methods to retrieve this information. However, the WEB is a highly volatile and heterogeneous information source, thus keyword based querying may not be the best approach when few information is given. This is due to the fact that different users with different needs may want distinct information, although related to the same keyword query. The process of relevance feedback makes it possible for the user to interact actively with the search engine. The main idea is that after performing an initial search in the WEB, the process enables the user to indicate, among the retrieved sites, a small number of the ones considered relevant or irrelevant according with his/her required information. The users preferences can then be used to rearrange sites returned in the initial search, so that relevant sites are ranked first. As in most cases a search returns a large amount of WEB sites which fits the keyword query, this is an ideal situation to use partially supervised machine learning algorithms. This kind of learning algorithms require a small number of labeled examples, and a large number of unlabeled examples. Thus, based on the assumption that the use of partially supervised learning is appropriate to induce a classifier that can be used as a filter for relevance feedback in WEB information retrieval, the aim of this work is to explore the use of a partially supervised machine learning algorithm, more specifically, one that uses multi-description data, in order to assist the WEB search. To this end, a computational tool called C-SEARCH, which performs the reordering of the searched results using the users feedback, has been implemented. Experimental results show that in cases where the keyword query is generic and there is a clear distinction between relevant and irrelevant sites, which is recognized by the user, the system can achieve good results

Dada a grande quantidade e diversidade de dados textuais sendo criados diariamente, as aplicações do processo de Mineração de Textos são inúmeras e variadas. Nesse processo, a qualidade da solução final depende, em parte, do modelo de representação de textos adotado. Por se tratar de textos em língua natural, relações sintáticas e semânticas influenciam o seu significado. No entanto, modelos tradicionais de representação de textos se limitam às palavras, não sendo possível diferenciar documentos que possuem o mesmo vocabulário, mas que apresentam visões diferentes sobre um mesmo assunto. Nesse contexto, este trabalho foi motivado pela diversidade das aplicações da tarefa de classificação automática de textos, pelo potencial das representações no modelo espaço-vetorial e pela lacuna referente ao tratamento da semântica inerente aos dados em língua natural. O seu desenvolvimento teve o propósito geral de avançar as pesquisas da área de Mineração de Textos em relação à incorporação de aspectos semânticos na representação de coleções de documentos. Um mapeamento sistemático da literatura da área foi realizado e os problemas de classificação foram categorizados em relação à complexidade semântica envolvida. Aspectos semânticos foram abordados com a proposta, bem como o desenvolvimento e a avaliação de sete modelos de representação de textos: (i) gBoED, modelo que incorpora a semântica obtida por meio de conhecimento do domínio; (ii) Uni-based, modelo que incorpora a semântica por meio da desambiguação lexical de sentidos e hiperônimos de conceitos; (iii) SR-based Terms e SR-based Sentences, modelos que incorporam a semântica por meio de anotações de papéis semânticos; (iv) NASARIdocs, Babel2Vec e NASARI+Babel2Vec, modelos que incorporam a semântica por meio de desambiguação lexical de sentidos e embeddings de palavras e conceitos. Representações de coleções de documentos geradas com os modelos propostos e outros da literatura foram analisadas e avaliadas na classificação automática de textos, considerando datasets de diferentes níveis de complexidade semântica. As propostas gBoED, Uni-based, SR-based Terms e SR-based Sentences apresentam atributos mais expressivos e possibilitam uma melhor interpretação da representação dos documentos. Já as propostas NASARIdocs, Babel2Vec e NASARI+Babel2Vec incorporam, de maneira latente, a semântica obtida de embeddings geradas a partir de uma grande quantidade de documentos externos. Essa propriedade tem um impacto positivo na performance de classificação. / Text Mining applications are numerous and varied since a huge amount of textual data are created daily. The quality of the final solution of a Text Mining process depends, among other factors, on the adopted text representation model. Despite the fact that syntactic and semantic relations influence natural language meaning, traditional text representation models are limited to words. The use of such models does not allow the differentiation of documents that use the same vocabulary but present different ideas about the same subject. The motivation of this work relies on the diversity of text classification applications, the potential of vector space model representations and the challenge of dealing with text semantics. Having the general purpose of advance the field of semantic representation of documents, we first conducted a systematic mapping study of semantics-concerned Text Mining studies and we categorized classification problems according to their semantic complexity. Then, we approached semantic aspects of texts through the proposal, analysis, and evaluation of seven text representation models: (i) gBoED, which incorporates text semantics by the use of domain expressions; (ii) Uni-based, which takes advantage of word sense disambiguation and hypernym relations; (iii) SR-based Terms and SR-based Sentences, which make use of semantic role labels; (iv) NASARIdocs, Babel2Vec and NASARI+Babel2Vec, which take advantage of word sense disambiguation and embeddings of words and senses.We analyzed the expressiveness and interpretability of the proposed text representation models and evaluated their classification performance against different literature models. While the proposed models gBoED, Uni-based, SR-based Terms and SR-based Sentences have improved expressiveness, the proposals NASARIdocs, Babel2Vec and NASARI+Babel2Vec are latently enriched by the embeddings semantics, obtained from the large training corpus. This property has a positive impact on text classification performance.

The task of opinion mining from product reviews has been achieved by employing rule-based approaches or generative learning models such as hidden Markov models (HMMs). This paper introduced a discriminative model using linear-chain Conditional Random Fields (CRFs) that can naturally incorporate arbitrary, non-independent features of the input without conditional independence among the features or distributional assumptions of inputs. The framework firstly performs part-of-speech (POS) tagging tasks over each word in sentences of review text. The performance is evaluated based on three criteria: precision, recall and F-score. The result shows that this approach is effective for this type of natural language processing (NLP) tasks. Then the framework extracts the keywords associated with each product feature and summarizes into concise lists that are simple and intuitive for people to read.

The elections in South Africa are contested by multiple political parties appealing to a diverse population that comes from a variety of socioeconomic backgrounds. As a result, a rich source of discourse is created to inform voters about election-related content. Two common sources of information to help voters with their decision are news articles and tweets, this study aims to understand the discourse in these two sources using natural language processing. Topic modelling techniques, Latent Dirichlet Allocation and Non- negative Matrix Factorization, are applied to digest the breadth of information collected about the elections into topics. The topics produced are subjected to further analysis that uncovers similarities between topics, links topics to dates and events and provides a summary of the discourse that existed prior to the South African general elections. The primary focus is on the 2019 elections, however election-related articles from 2014 and 2019 were also compared to understand how the discourse has changed. / Mini Dissertation (MIT (Big Data Science))--University of Pretoria, 2019. / Computer Science / MIT (Big Data Science) / Unrestricted

This master’s thesis deals with using Text mining as a method to predict tags of articles. It describes the iterative way of handling big data files, parsing the data, cleaning the data and scoring of terms in article using TF-IDF. It describes in detail the flow of program written in programming language Python 3.4.3. The result of processing more than 1 million articles from Wikipedia database is a dictionary of English terms. By using this dictionary one is capable of determining the most important terms from article in corpus of articles. Relevancy of consequent tags proves the method used in this case.

This thesis deals with text mining. It focuses on problems of document classification and related techniques, mainly data preprocessing. Project also introduces the SVM method, which has been chosen for classification, design and testing of implemented application.

As information and communication technology develops swiftly, amount of information produced by various sources grows as well. Sorting and obtaining knowledge from this data requires significant effort which is not ensured easily by a human, meaning machine processing is taking place. Acquiring emotion from text data is an interesting area of research and it’s going through considerable expansion while being used widely. Purpose of this thesis is to create a system for positive and negative emotion detection from text along with evaluation of its performance. System was created with Java programming language and it allows training with use of large amount of data (known as Big Data), exploiting Spark library. Thesis describes structure and handling text from database used as source of input data. Classificator model was created with use of Support Vector Machines and optimized by the n-grams method.

La prolifération des données numériques a permis aux communautés de scientifiques et de praticiens de créer de nouvelles technologies basées sur les données pour mieux connaître les utilisateurs finaux et en particulier leur comportement. L’objectif est alors de fournir de meilleurs services et un meilleur support aux personnes dans leur expérience numérique. La majorité de ces technologies créées pour analyser le comportement humain utilisent très souvent des données de logs générées passivement au cours de l’interaction homme-machine. Une particularité de ces traces comportementales est qu’elles sont enregistrées et stockées selon une structure clairement définie. En revanche, les traces générées de manière proactive sont très peu structurées et représentent la grande majorité des données numériques existantes. De plus, les données non structurées se trouvent principalement sous forme de texte. À ce jour, malgré la prédominance des données textuelles et la pertinence des connaissances comportementales dans de nombreux domaines, les textes numériques sont encore insuffisamment étudiés en tant que traces du comportement humain pour révéler automatiquement des connaissances détaillées sur le comportement.L’objectif de recherche de cette thèse est de proposer une méthode indépendante du corpus pour exploiter automatiquement les communications asynchrones en tant que traces de comportement générées de manière proactive afin de découvrir des modèles de processus de conversations,axés sur des intentions de discours et des relations, toutes deux exhaustives et détaillées.Plusieurs contributions originales sont faites. Il y est menée la seule revue systématique existante à ce jour sur la modélisation automatique des conversations asynchrones avec des actes de langage. Une taxonomie des intentions de discours est dérivée de la linguistique pour modéliser la communication asynchrone. Comparée à toutes les taxonomies des travaux connexes,celle proposée est indépendante du corpus, à la fois plus détaillée et exhaustive dans le contexte donné, et son application par des non-experts est prouvée au travers d’expériences approfondies.Une méthode automatique, indépendante du corpus, pour annoter les énoncées de communication asynchrone avec la taxonomie des intentions de discours proposée, est conçue sur la base d’un apprentissage automatique supervisé. Pour cela, deux corpus "ground-truth" validés sont créés et trois groupes de caractéristiques (discours, contenu et conversation) sont conçus pour être utilisés par les classificateurs. En particulier, certaines des caractéristiques du discours sont nouvelles et définies en considérant des moyens linguistiques pour exprimer des intentions de discours,sans s’appuyer sur le contenu explicite du corpus, le domaine ou les spécificités des types de communication asynchrones. Une méthode automatique basée sur la fouille de processus est conçue pour générer des modèles de processus d’intentions de discours interdépendantes à partir de tours de parole, annotés avec plusieurs labels par phrase. Comme la fouille de processus repose sur des logs d’événements structurés et bien définis, un algorithme est proposé pour produire de tels logs d’événements à partir de conversations. Par ailleurs, d’autres solutions pour transformer les conversations annotées avec plusieurs labels par phrase en logs d’événements, ainsi que l’impact des différentes décisions sur les modèles comportementaux en sortie sont analysées afin d’alimenter de futures recherches.Des expériences et des validations qualitatives à la fois en médecine et en analyse conversationnelle montrent que la solution proposée donne des résultats fiables et pertinents. Cependant,des limitations sont également identifiées, elles devront être abordées dans de futurs travaux. / The proliferation of digital data has enabled scientific and practitioner communities to createnew data-driven technologies to learn about user behaviors in order to deliver better services and support to people in their digital experience. The majority of these technologies extensively derive value from data logs passively generated during the human-computer interaction. A particularity of these behavioral traces is that they are structured. However, the pro-actively generated text across Internet is highly unstructured and represents the overwhelming majority of behavioral traces. To date, despite its prevalence and the relevance of behavioral knowledge to many domains, such as recommender systems, cyber-security and social network analysis,the digital text is still insufficiently tackled as traces of human behavior to automatically reveal extensive insights into behavior.The main objective of this thesis is to propose a corpus-independent method to automatically exploit the asynchronous communication as pro-actively generated behavior traces in order to discover process models of conversations, centered on comprehensive speech intentions and relations. The solution is built in three iterations, following a design science approach.Multiple original contributions are made. The only systematic study to date on the automatic modeling of asynchronous communication with speech intentions is conducted. A speech intention taxonomy is derived from linguistics to model the asynchronous communication and, comparedto all taxonomies from the related works, it is corpus-independent, comprehensive—as in both finer-grained and exhaustive in the given context, and its application by non-experts is proven feasible through extensive experiments. A corpus-independent, automatic method to annotate utterances of asynchronous communication with the proposed speech intention taxonomy is designed based on supervised machine learning. For this, validated ground-truth corpora arecreated and groups of features—discourse, content and conversation-related, are engineered to be used by the classifiers. In particular, some of the discourse features are novel and defined by considering linguistic means to express speech intentions, without relying on the corpus explicit content, domain or on specificities of the asynchronous communication types. Then, an automatic method based on process mining is designed to generate process models of interrelated speech intentions from conversation turns, annotated with multiple speech intentions per sentence. As process mining relies on well-defined structured event logs, an algorithm to produce such logs from conversations is proposed. Additionally, an extensive design rationale on how conversations annotated with multiple labels per sentence could be transformed in event logs and what is the impact of different decisions on the output behavioral models is released to support future research. Experiments and qualitative validations in medicine and conversation analysis show that the proposed solution reveals reliable and relevant results, but also limitations are identified,to be addressed in future works.

In this report we investigate how much similarity good news and bad news may have in context of long-terms market trends. We discuss the relation between text mining, classification, and information retrieval. We present examples that use identical set of words but have a quite different meaning, we present examples that can be interpreted in both positive or negative sense so that the decision is difficult as before reading them. Our examples prove that methods of information retrieval are not strong enough to solve problems as specified above. For searching of common properties in groups of news we had used classifiers (e.g. naive Bayes classifier) after we found that the use of diagnostic methods did not deliver reasonable results. For our experiments we have used historical data concerning the German market index DAX 30. / In diesem Bericht untersuchen wir, wieviel Ähnlichkeit gute und schlechte Nachrichten im Kontext von Langzeitmarkttrends besitzen. Wir diskutieren die Verbindungen zwischen Text Mining, Klassifikation und Information Retrieval. Wir präsentieren Beispiele, die identische Wortmengen verwenden, aber trotzdem recht unterschiedliche Bedeutungen besitzen; Beispiele, die sowohl positiv als auch negativ interpretiert werden können. Sie zeigen Probleme auf, die mit Methoden des Information Retrieval nicht gelöst werden können. Um nach Gemeinsamkeiten in Nachrichtengruppen zu suchen, verwendeten wir Klassifikatoren (z.B. Naive Bayes), nachdem wir herausgefunden hatten, dass der Einsatz von diagnostizierenden Methoden keine vernünftigen Resultate erzielte. Für unsere Experimente nutzten wir historische Daten des Deutschen Aktienindex DAX 30.

The advent of Web 2.0 technologies which supports the creation and publishing of various social media content in a collaborative and participatory way by all users in the form of user generated content and social networks has led to the creation of vast amounts of structured, semi-structured and unstructured data. The sudden rise of social media has led to their wide adoption by organisations of various sizes worldwide in order to take advantage of this new way of communication and engaging with their stakeholders in ways that was unimaginable before. Data generated from social media is highly unstructured, which makes it challenging for most organisations which are normally used for handling and analysing structured data from business transactions. The research reported in this dissertation was carried out to investigate fast and efficient methods available for retrieving, storing and analysing unstructured data form social media in order to make crucial and informed business decisions on time. Sentiment analysis was conducted on Twitter data called tweets. Twitter, which is one of the most widely adopted social network service provides an API (Application Programming Interface), for researchers and software developers to connect and collect public data sets of Twitter data from the Twitter database. A Twitter application was created and used to collect streams of real-time public data via a Twitter source provided by Apache Flume and efficiently storing this data in Hadoop File System (HDFS). Apache Flume is a distributed, reliable, and available system which is used to efficiently collect, aggregate and move large amounts of log data from many different sources to a centralized data store such as HDFS. Apache Hadoop is an open source software library that runs on low-cost commodity hardware and has the ability to store, manage and analyse large amounts of both structured and unstructured data quickly, reliably, and flexibly at low-cost. A Lexicon based sentiment analysis approach was taken and the AFINN-111 lexicon was used for scoring. The Twitter data was analysed from the HDFS using a Java MapReduce implementation. MapReduce is a programming model and an associated implementation for processing and generating big data sets with a parallel, distributed algorithm on a cluster. The results demonstrate that it is fast, efficient and economical to use this approach to analyse unstructured data from social media in real time. / Dissertation (MSc)--University of Pretoria, 2019. / National Research Foundation (NRF) - Scarce skills / Computer Science / MSc / Unrestricted