In this paper, a simple text categorization method using term-class relevance measures is proposed. Initially, text documents are processed to extract significant terms present in them. For every term extracted from a document, we compute its importance in preserving the content of a class through a novel term-weighting scheme known as Term_Class Relevance (TCR) measure proposed by Guru and Suhil (2015) [1]. In this way, for every term, its relevance for all the classes present in the corpus is computed and stored in the knowledgebase. During testing, the terms present in the test document are extracted and the term-class relevance of each term is obtained from the stored knowledgebase. To achieve quick search of term weights, Btree indexing data structure has been adapted. Finally, the class which receives maximum support in terms of term-class relevance is decided to be the class of the given test document. The proposed method works in logarithmic complexity in testing time and simple to implement when compared to any other text categorization techniques available in literature. The experiments conducted on various benchmarking datasets have revealed that the performance of the proposed method is satisfactory and encouraging.
Bayesian max-margin models have shown superiority in various practical applications, such as text categorization, collaborative prediction, social network link prediction and crowdsourcing, and they conjoin the flexibility of Bayesian modeling and predictive strengths of max-margin learning. However, Monte Carlo sampling for these models still remains challenging, especially for applications that involve large-scale datasets. In this paper, we present the stochastic subgradient Hamiltonian Monte Carlo (HMC) methods, which are easy to implement and computationally efficient. We show the approximate detailed balance property of subgradient HMC which reveals a natural and validated generalization of the ordinary HMC. Furthermore, we investigate the variants that use stochastic subsampling and thermostats for better scalability and mixing. Using stochastic subgradient Markov Chain Monte Carlo (MCMC), we efficiently solve the posterior inference task of various Bayesian max-margin models and extensive experimental results demonstrate the effectiveness of our approach.
Progressive filtering is a simple way to perform hierarchical classification, inspired by the behavior that most humans put into practice while attempting to categorize an item according to an underlying taxonomy. Each node of the taxonomy being associated with a different category, one may visualize the categorization process by looking at the item going downwards through all the nodes that accept it as belonging to the corresponding category. This paper is aimed at modeling the progressive filtering technique from a probabilistic perspective, in a hierarchical text categorization setting. As a result, the designer of a system based on progressive filtering should be facilitated in the task of devising, training, and testing it.
This paper presents an approach to classify documents in any language into an English topical label space, without any text categorization training data. The approach, Cross-Lingual Dataless Document Classification (CLDDC) relies on mapping the English labels or short category description into a Wikipedia-based semantic representation, and on the use of the target language Wikipedia. Consequently, performance could suffer when Wikipedia in the target language is small. In this paper, we focus on languages with small Wikipedias, (Small-Wikipedia languages, SWLs). We use a word-level dictionary to convert documents in a SWL to a large-Wikipedia language (LWLs), and then perform CLDDC based on the LWL's Wikipedia. This approach can be applied to thousands of languages, which can be contrasted with machine translation, which is a supervision heavy approach and can be done for about 100 languages. We also develop a ranking algorithm that makes use of language similarity metrics to automatically select a good LWL, and show that this significantly improves classification of SWLs' documents, performing comparably to the best bridge possible.
Turkish Wikipedia Named-Entity Recognition and Text Categorization (TWNERTC) dataset is a collection of automatically categorized and annotated sentences obtained from Wikipedia. We constructed large-scale gazetteers by using a graph crawler algorithm to extract relevant entity and domain information from a semantic knowledge base, Freebase. The constructed gazetteers contains approximately 300K entities with thousands of fine-grained entity types under 77 different domains. Since automated processes are prone to ambiguity, we also introduce two new content specific noise reduction methodologies. Moreover, we map fine-grained entity types to the equivalent four coarse-grained types: person, loc, org, misc. Eventually, we construct six different dataset versions and evaluate the quality of annotations by comparing ground truths from human annotators. We make these datasets publicly available to support studies on Turkish named-entity recognition (NER) and text categorization (TC).
The Entity Disambiguation and Linking (EDL) task matches entity mentions in text to a unique Knowledge Base (KB) identifier such as a Wikipedia or Freebase id. It plays a critical role in the construction of a high quality information network, and can be further leveraged for a variety of information retrieval and NLP tasks such as text categorization and document tagging. EDL is a complex and challenging problem due to ambiguity of the mentions and real world text being multi-lingual. Moreover, EDL systems need to have high throughput and should be lightweight in order to scale to large datasets and run on off-the-shelf machines. More importantly, these systems need to be able to extract and disambiguate dense annotations from the data in order to enable an Information Retrieval or Extraction task running on the data to be more efficient and accurate. In order to address all these challenges, we present the Lithium EDL system and algorithm - a high-throughput, lightweight, language-agnostic EDL system that extracts and correctly disambiguates 75% more entities than state-of-the-art EDL systems and is significantly faster than them.
A fundamental advantage of neural models for NLP is their ability to learn representations from scratch. However, in practice this often means ignoring existing external linguistic resources, e.g., WordNet or domain specific ontologies such as the Unified Medical Language System (UMLS). We propose a general, novel method for exploiting such resources via weight sharing. Prior work on weight sharing in neural networks has considered it largely as a means of model compression. In contrast, we treat weight sharing as a flexible mechanism for incorporating prior knowledge into neural models. We show that this approach consistently yields improved performance on classification tasks compared to baseline strategies that do not exploit weight sharing.
We show that discourse structure, as defined by Rhetorical Structure Theory and provided by an existing discourse parser, benefits text categorization. Our approach uses a recursive neural network and a newly proposed attention mechanism to compute a representation of the text that focuses on salient content, from the perspective of both RST and the task. Experiments consider variants of the approach and illustrate its strengths and weaknesses.
JaTeCS is an open source Java library that supports research on automatic text categorization and other related problems, such as ordinal regression and quantification, which are of special interest in opinion mining applications. It covers all the steps of an experimental activity, from reading the corpus to the evaluation of the experimental results. As JaTeCS is focused on text as the main input data, it provides the user with many text-dedicated tools, e.g.: data readers for many formats, including the most commonly used text corpora and lexical resources, natural language processing tools, multi-language support, methods for feature selection and weighting, the implementation of many machine learning algorithms as well as wrappers for well-known external software (e.g., SVM_light) which enable their full control from code. JaTeCS support its expansion by abstracting through interfaces many of the typical tools and procedures used in text processing tasks. The library also provides a number of "template" implementations of typical experimental setups (e.g., train-test, k-fold validation, grid-search optimization, randomized runs) which enable fast realization of experiments just by connecting the templates with data readers, learning algorithms and evaluation measures.
In this paper, we present a semi-supervised learning algorithm for classification of text documents. A method of labeling unlabeled text documents is presented. The presented method is based on the principle of divide and conquer strategy. It uses recursive K-means algorithm for partitioning both labeled and unlabeled data collection. The K-means algorithm is applied recursively on each partition till a desired level partition is achieved such that each partition contains labeled documents of a single class. Once the desired clusters are obtained, the respective cluster centroids are considered as representatives of the clusters and the nearest neighbor rule is used for classifying an unknown text document. Series of experiments have been conducted to bring out the superiority of the proposed model over other recent state of the art models on 20Newsgroups dataset.
In this work, a problem associated with imbalanced text corpora is addressed. A method of converting an imbalanced text corpus into a balanced one is presented. The presented method employs a clustering algorithm for conversion. Initially to avoid curse of dimensionality, an effective representation scheme based on term class relevancy measure is adapted, which drastically reduces the dimension to the number of classes in the corpus. Subsequently, the samples of larger sized classes are grouped into a number of subclasses of smaller sizes to make the entire corpus balanced. Each subclass is then given a single symbolic vector representation by the use of interval valued features. This symbolic representation in addition to being compact helps in reducing the space requirement and also the classification time. The proposed model has been empirically demonstrated for its superiority on bench marking datasets viz., Reuters 21578 and TDT2. Further, it has been compared against several other existing contemporary models including model based on support vector machine. The comparative analysis indicates that the proposed model outperforms the other existing models.
Text preprocessing is often the first step in the pipeline of a Natural Language Processing (NLP) system, with potential impact in its final performance. Despite its importance, text preprocessing has not received much attention in the deep learning literature. In this paper we investigate the impact of simple text preprocessing decisions (particularly tokenizing, lemmatizing, lowercasing and multiword grouping) on the performance of a standard neural text classifier. We perform an extensive evaluation on standard benchmarks from text categorization and sentiment analysis. While our experiments show that a simple tokenization of input text is generally adequate, they also highlight significant degrees of variability across preprocessing techniques. This reveals the importance of paying attention to this usually-overlooked step in the pipeline, particularly when comparing different models. Finally, our evaluation provides insights into the best preprocessing practices for training word embeddings.
In this paper, we propose a novel approach for text classification based on clustering word embeddings, inspired by the bag of visual words model, which is widely used in computer vision. After each word in a collection of documents is represented as word vector using a pre-trained word embeddings model, a k-means algorithm is applied on the word vectors in order to obtain a fixed-size set of clusters. The centroid of each cluster is interpreted as a super word embedding that embodies all the semantically related word vectors in a certain region of the embedding space. Every embedded word in the collection of documents is then assigned to the nearest cluster centroid. In the end, each document is represented as a bag of super word embeddings by computing the frequency of each super word embedding in the respective document. We also diverge from the idea of building a single vocabulary for the entire collection of documents, and propose to build class-specific vocabularies for better performance. Using this kind of representation, we report results on two text mining tasks, namely text categorization by topic and polarity classification. On both tasks, our model yields better performance than the standard bag of words.
Sarcasm detection is a key task for many natural language processing tasks. In sentiment analysis, for example, sarcasm can flip the polarity of an "apparently positive" sentence and, hence, negatively affect polarity detection performance. To date, most approaches to sarcasm detection have treated the task primarily as a text categorization problem. Sarcasm, however, can be expressed in very subtle ways and requires a deeper understanding of natural language that standard text categorization techniques cannot grasp. In this work, we develop models based on a pre-trained convolutional neural network for extracting sentiment, emotion and personality features for sarcasm detection. Such features, along with the network's baseline features, allow the proposed models to outperform the state of the art on benchmark datasets. We also address the often ignored generalizability issue of classifying data that have not been seen by the models at learning phase.
In this paper, we discuss a different type of semi-supervised setting: a coarse level of labeling is available for all observations but the model has to learn a fine level of latent annotation for each one of them. Problems in this setting are likely to be encountered in many domains such as text categorization, protein function prediction, image classification as well as in exploratory scientific studies such as medical and genomics research. We consider this setting as simultaneously performed supervised classification (per the available coarse labels) and unsupervised clustering (within each one of the coarse labels) and propose a novel output layer modification called auto-clustering output layer (ACOL) that allows concurrent classification and clustering based on Graph-based Activity Regularization (GAR) technique. As the proposed output layer modification duplicates the softmax nodes at the output layer for each class, GAR allows for competitive learning between these duplicates on a traditional error-correction learning framework to ultimately enable a neural network to learn the latent annotations in this partially supervised setup. We demonstrate how the coarse label supervision impacts performance and helps propagate useful clustering information between sub-classes. Comparative tests on three of the most popular image datasets MNIST, SVHN and CIFAR-100 rigorously demonstrate the effectiveness and competitiveness of the proposed approach.
A great variety of text tasks such as topic or spam identification, user profiling, and sentiment analysis can be posed as a supervised learning problem and tackle using a text classifier. A text classifier consists of several subprocesses, some of them are general enough to be applied to any supervised learning problem, whereas others are specifically designed to tackle a particular task, using complex and computational expensive processes such as lemmatization, syntactic analysis, etc. Contrary to traditional approaches, we propose a minimalistic and wide system able to tackle text classification tasks independent of domain and language, namely microTC. It is composed by some easy to implement text transformations, text representations, and a supervised learning algorithm. These pieces produce a competitive classifier even in the domain of informally written text. We provide a detailed description of microTC along with an extensive experimental comparison with relevant state-of-the-art methods. mircoTC was compared on 30 different datasets. Regarding accuracy, microTC obtained the best performance in 20 datasets while achieves competitive results in the remaining 10. The compared datasets include several problems like topic and polarity classification, spam detection, user profiling and authorship attribution. Furthermore, it is important to state that our approach allows the usage of the technology even without knowledge of machine learning and natural language processing.
This contribution to a special issue on "Computer-aided processing of intertextuality" in ancient texts will illustrate how using digital tools to interact with the Hebrew Bible offers new promising perspectives for visualizing the texts and for performing tasks in education and research. This contribution explores how the corpus of the Hebrew Bible created and maintained by the Eep Talstra Centre for Bible and Computer can support new methods for modern knowledge workers within the field of digital humanities and theology be applied to ancient texts, and how this can be envisioned as a new field of digital intertextuality. The article first describes how the corpus was used to develop the Bible Online Learner as a persuasive technology to enhance language learning with, in, and around a database that acts as the engine driving interactive tasks for learners. Intertextuality in this case is a matter of active exploration and ongoing practice. Furthermore, interactive corpus-technology has an important bearing on the task of textual criticism as a specialized area of research that depends increasingly on the availability of digital resources. Commercial solutions developed by software companies like Logos and Accordance offer a market-based intertextuality defined by the production of advanced digital resources for scholars and students as useful alternatives to often inaccessible and expensive printed versions. It is reasonable to expect that in the future interactive corpus technology will allow scholars to do innovative academic tasks in textual criticism and interpretation. We have already seen the emergence of promising tools for text categorization, analysis of translation shifts, and interpretation. Broadly speaking, interactive tools and tasks within the three areas of language learning, textual criticism, and Biblical studies illustrate a new kind of intertextuality emerging within digital humanities.
This paper investigates, from information theoretic grounds, a learning problem based on the principle that any regularity in a given dataset can be exploited to extract compact features from data, i.e., using fewer bits than needed to fully describe the data itself, in order to build meaningful representations of a relevant content (multiple labels). We begin by introducing the noisy lossy source coding paradigm with the log-loss fidelity criterion which provides the fundamental tradeoffs between the \emph{cross-entropy loss} (average risk) and the information rate of the features (model complexity). Our approach allows an information theoretic formulation of the \emph{multi-task learning} (MTL) problem which is a supervised learning framework in which the prediction models for several related tasks are learned jointly from common representations to achieve better generalization performance. Then, we present an iterative algorithm for computing the optimal tradeoffs and its global convergence is proven provided that some conditions hold. An important property of this algorithm is that it provides a natural safeguard against overfitting, because it minimizes the average risk taking into account a penalization induced by the model complexity. Remarkably, empirical results illustrate that there exists an optimal information rate minimizing the \emph{excess risk} which depends on the nature and the amount of available training data. An application to hierarchical text categorization is also investigated, extending previous works.
Unsupervised feature extractors are known to perform an efficient and discriminative representation of data. Insight into the mappings they perform and human ability to understand them, however, remain very limited. This is especially prominent when multilayer deep learning architectures are used. This paper demonstrates how to remove these bottlenecks within the architecture of Nonnegativity Constrained Autoencoder (NCSAE). It is shown that by using both L1 and L2 regularization that induce nonnegativity of weights, most of the weights in the network become constrained to be nonnegative thereby resulting into a more understandable structure with minute deterioration in classification accuracy. Also, this proposed approach extracts features that are more sparse and produces additional output layer sparsification. The method is analyzed for accuracy and feature interpretation on the MNIST data, the NORB normalized uniform object data, and the Reuters text categorization dataset.
Many applications require categorization of text documents using predefined categories. The main approach to performing text categorization is learning from labeled examples. For many tasks, it may be difficult to find examples in one language but easy in others. The problem of learning from examples in one or more languages and classifying (categorizing) in another is called cross-lingual learning. In this work, we present a novel approach that solves the general cross-lingual text categorization problem. Our method generates, for each training document, a set of language-independent features. Using these features for training yields a language-independent classifier. At the classification stage, we generate language-independent features for the unlabeled document, and apply the classifier on the new representation. To build the feature generator, we utilize a hierarchical language-independent ontology, where each concept has a set of support documents for each language involved. In the preprocessing stage, we use the support documents to build a set of language-independent feature generators, one for each language. The collection of these generators is used to map any document into the language-independent feature space. Our methodology works on the most general cross-lingual text categorization problems, being able to learn from any mix of languages and classify documents in any other language. We also present a method for exploiting the hierarchical structure of the ontology to create virtual supporting documents for languages that do not have them. We tested our method, using Wikipedia as our ontology, on the most commonly used test collections in cross-lingual text categorization, and found that it outperforms existing methods.