Question answering is an important and difficult task in the natural language processing domain, because many basic natural language processing tasks can be cast into a question answering task. Several deep neural network architectures have been developed recently, which employ memory and inference components to memorize and reason over text information, and generate answers to questions. However, a major drawback of many such models is that they are capable of only generating single-word answers. In addition, they require large amount of training data to generate accurate answers. In this paper, we introduce the Long-Term Memory Network (LTMN), which incorporates both an external memory module and a Long Short-Term Memory (LSTM) module to comprehend the input data and generate multi-word answers. The LTMN model can be trained end-to-end using back-propagation and requires minimal supervision. We test our model on two synthetic data sets (based on Facebook's bAbI data set) and the real-world Stanford question answering data set, and show that it can achieve state-of-the-art performance.
Building recommendation algorithms is one of the most challenging tasks in Machine Learning. Although most of the recommendation systems are built on explicit feedback available from the users in terms of rating or text, a majority of the applications do not receive such feedback. Here we consider the recommendation task where the only available data is the records of user-item interaction over web applications over time, in terms of subscription or purchase of items; this is known as implicit feedback recommendation. There is usually a massive amount of such user-item interaction available for any web applications. Algorithms like PLSI or Matrix Factorization runs several iterations through the dataset, and may prove very expensive for large datasets. Here we propose a recommendation algorithm based on Method of Moment, which involves factorization of second and third order moments of the dataset. Our algorithm can be proven to be globally convergent using PAC learning theory. Further, we show how to extract the parameters using only three passes through the entire dataset. This results in a highly scalable algorithm that scales up to million of users even on a machine with a single-core processor and 8 GB RAM and produces competitive performance in comparison with existing algorithms.
Modern applications and progress in deep learning research have created renewed interest for generative models of text and of images. However, even today it is unclear what objective functions one should use to train and evaluate these models. In this paper we present two contributions. Firstly, we present a critique of scheduled sampling, a state-of-the-art training method that contributed to the winning entry to the MSCOCO image captioning benchmark in 2015. Here we show that despite this impressive empirical performance, the objective function underlying scheduled sampling is improper and leads to an inconsistent learning algorithm. Secondly, we revisit the problems that scheduled sampling was meant to address, and present an alternative interpretation. We argue that maximum likelihood is an inappropriate training objective when the end-goal is to generate natural-looking samples. We go on to derive an ideal objective function to use in this situation instead. We introduce a generalisation of adversarial training, and show how such method can interpolate between maximum likelihood training and our ideal training objective. To our knowledge this is the first theoretical analysis that explains why adversarial training tends to produce samples with higher perceived quality.
Bidirectional recurrent neural networks (RNN) are trained to predict both in the positive and negative time directions simultaneously. They have not been used commonly in unsupervised tasks, because a probabilistic interpretation of the model has been difficult. Recently, two different frameworks, GSN and NADE, provide a connection between reconstruction and probabilistic modeling, which makes the interpretation possible. As far as we know, neither GSN or NADE have been studied in the context of time series before. As an example of an unsupervised task, we study the problem of filling in gaps in high-dimensional time series with complex dynamics. Although unidirectional RNNs have recently been trained successfully to model such time series, inference in the negative time direction is non-trivial. We propose two probabilistic interpretations of bidirectional RNNs that can be used to reconstruct missing gaps efficiently. Our experiments on text data show that both proposed methods are much more accurate than unidirectional reconstructions, although a bit less accurate than a computationally complex bidirectional Bayesian inference on the unidirectional RNN. We also provide results on music data for which the Bayesian inference is computationally infeasible, demonstrating the scalability of the proposed methods.
Recent progress on automatic generation of image captions has shown that it is possible to describe the most salient information conveyed by images with accurate and meaningful sentences. In this paper, we propose an image caption system that exploits the parallel structures between images and sentences. In our model, the process of generating the next word, given the previously generated ones, is aligned with the visual perception experience where the attention shifting among the visual regions imposes a thread of visual ordering. This alignment characterizes the flow of "abstract meaning", encoding what is semantically shared by both the visual scene and the text description. Our system also makes another novel modeling contribution by introducing scene-specific contexts that capture higher-level semantic information encoded in an image. The contexts adapt language models for word generation to specific scene types. We benchmark our system and contrast to published results on several popular datasets. We show that using either region-based attention or scene-specific contexts improves systems without those components. Furthermore, combining these two modeling ingredients attains the state-of-the-art performance.
Owing to the rapidly growing multimedia content available on the Internet, extractive spoken document summarization, with the purpose of automatically selecting a set of representative sentences from a spoken document to concisely express the most important theme of the document, has been an active area of research and experimentation. On the other hand, word embedding has emerged as a newly favorite research subject because of its excellent performance in many natural language processing (NLP)-related tasks. However, as far as we are aware, there are relatively few studies investigating its use in extractive text or speech summarization. A common thread of leveraging word embeddings in the summarization process is to represent the document (or sentence) by averaging the word embeddings of the words occurring in the document (or sentence). Then, intuitively, the cosine similarity measure can be employed to determine the relevance degree between a pair of representations. Beyond the continued efforts made to improve the representation of words, this paper focuses on building novel and efficient ranking models based on the general word embedding methods for extractive speech summarization. Experimental results demonstrate the effectiveness of our proposed methods, compared to existing state-of-the-art methods.
Searching through a large volume of data is very critical for companies, scientists, and searching engines applications due to time complexity and memory complexity. In this paper, a new technique of generating FuzzyFind Dictionary for text mining was introduced. We simply mapped the 23 bits of the English alphabet into a FuzzyFind Dictionary or more than 23 bits by using more FuzzyFind Dictionary, and reflecting the presence or absence of particular letters. This representation preserves closeness of word distortions in terms of closeness of the created binary vectors within Hamming distance of 2 deviations. This paper talks about the Golay Coding Transformation Hash Table and how it can be used on a FuzzyFind Dictionary as a new technology for using in searching through big data. This method is introduced by linear time complexity for generating the dictionary and constant time complexity to access the data and update by new data sets, also updating for new data sets is linear time depends on new data points. This technique is based on searching only for letters of English that each segment has 23 bits, and also we have more than 23-bit and also it could work with more segments as reference table.
Neural network techniques are widely applied to obtain high-quality distributed representations of words, i.e., word embeddings, to address text mining, information retrieval, and natural language processing tasks. Recently, efficient methods have been proposed to learn word embeddings from context that captures both semantic and syntactic relationships between words. However, it is challenging to handle unseen words or rare words with insufficient context. In this paper, inspired by the study on word recognition process in cognitive psychology, we propose to take advantage of seemingly less obvious but essentially important morphological knowledge to address these challenges. In particular, we introduce a novel neural network architecture called KNET that leverages both contextual information and morphological word similarity built based on morphological knowledge to learn word embeddings. Meanwhile, the learning architecture is also able to refine the pre-defined morphological knowledge and obtain more accurate word similarity. Experiments on an analogical reasoning task and a word similarity task both demonstrate that the proposed KNET framework can greatly enhance the effectiveness of word embeddings.
This paper describes a weakly supervised system for sentiment analysis in the movie review domain. The objective is to classify a movie review into a polarity class, positive or negative, based on those sentences bearing opinion on the movie alone. The irrelevant text, not directly related to the reviewer opinion on the movie, is left out of analysis. Wikipedia incorporates the world knowledge of movie-specific features in the system which is used to obtain an extractive summary of the review, consisting of the reviewer's opinions about the specific aspects of the movie. This filters out the concepts which are irrelevant or objective with respect to the given movie. The proposed system, WikiSent, does not require any labeled data for training. The only weak supervision arises out of the usage of resources like WordNet, Part-of-Speech Tagger and Sentiment Lexicons by virtue of their construction. WikiSent achieves a considerable accuracy improvement over the baseline and has a better or comparable accuracy to the existing semi-supervised and unsupervised systems in the domain, on the same dataset. We also perform a general movie review trend analysis using WikiSent to find the trend in movie-making and the public acceptance in terms of movie genre, year of release and polarity.
In this paper we present statistical analysis of English texts from Wikipedia. We try to address the issue of language complexity empirically by comparing the simple English Wikipedia (Simple) to comparable samples of the main English Wikipedia (Main). Simple is supposed to use a more simplified language with a limited vocabulary, and editors are explicitly requested to follow this guideline, yet in practice the vocabulary richness of both samples are at the same level. Detailed analysis of longer units (n-grams of words and part of speech tags) shows that the language of Simple is less complex than that of Main primarily due to the use of shorter sentences, as opposed to drastically simplified syntax or vocabulary. Comparing the two language varieties by the Gunning readability index supports this conclusion. We also report on the topical dependence of language complexity, e.g. that the language is more advanced in conceptual articles compared to person-based (biographical) and object-based articles. Finally, we investigate the relation between conflict and language complexity by analyzing the content of the talk pages associated to controversial and peacefully developing articles, concluding that controversy has the effect of reducing language complexity.