This paper presents machine learning solutions to a practical problem of Natural Language Generation (NLG), particularly the word formation in agglutinative languages like Tamil, in a supervised manner. The morphological generator is an important component of Natural Language Processing in Artificial Intelligence. It generates word forms given a root and affixes. The morphophonemic changes like addition, deletion, alternation etc., occur when two or more morphemes or words joined together. The Sandhi rules should be explicitly specified in the rule based morphological analyzers and generators. In machine learning framework, these rules can be learned automatically by the system from the training samples and subsequently be applied for new inputs. In this paper we proposed the machine learning models which learn the morphophonemic rules for noun declensions from the given training data. These models are trained to learn sandhi rules using various learning algorithms and the performance of those algorithms are presented. From this we conclude that machine learning of morphological processing such as word form generation can be successfully learned in a supervised manner, without explicit description of rules. The performance of Decision trees and Bayesian machine learning algorithms on noun declensions are discussed.
We present NaturalOWL, a natural language generation system that produces texts describing individuals or classes of OWL ontologies. Unlike simpler OWL verbalizers, which typically express a single axiom at a time in controlled, often not entirely fluent natural language primarily for the benefit of domain experts, we aim to generate fluent and coherent multi-sentence texts for end-users. With a system like NaturalOWL, one can publish information in OWL on the Web, along with automatically produced corresponding texts in multiple languages, making the information accessible not only to computer programs and domain experts, but also end-users. We discuss the processing stages of NaturalOWL, the optional domain-dependent linguistic resources that the system can use at each stage, and why they are useful. We also present trials showing that when the domain-dependent llinguistic resources are available, NaturalOWL produces significantly better texts compared to a simpler verbalizer, and that the resources can be created with relatively light effort.
In this paper we propose a general framework for learning distributed representations of attributes: characteristics of text whose representations can be jointly learned with word embeddings. Attributes can correspond to document indicators (to learn sentence vectors), language indicators (to learn distributed language representations), meta-data and side information (such as the age, gender and industry of a blogger) or representations of authors. We describe a third-order model where word context and attribute vectors interact multiplicatively to predict the next word in a sequence. This leads to the notion of conditional word similarity: how meanings of words change when conditioned on different attributes. We perform several experimental tasks including sentiment classification, cross-lingual document classification, and blog authorship attribution. We also qualitatively evaluate conditional word neighbours and attribute-conditioned text generation.
This paper presents a currently bilingual but potentially multilingual FrameNet-based grammar library implemented in Grammatical Framework. The contribution of this paper is two-fold. First, it offers a methodological approach to automatically generate the grammar based on semantico-syntactic valence patterns extracted from FrameNet-annotated corpora. Second, it provides a proof of concept for two use cases illustrating how the acquired multilingual grammar can be exploited in different CNL applications in the domains of arts and tourism.
An essential element of any verification technique is that of identifying and communicating to the user, system behaviour which leads to a deviation from the expected behaviour. Such behaviours are typically made available as long traces of system actions which would benefit from a natural language explanation of the trace and especially in the context of business logic level specifications. In this paper we present a natural language generation model which can be used to explain such traces. A key idea is that the explanation language is a CNL that is, formally speaking, regular language susceptible transformations that can be expressed with finite state machinery. At the same time it admits various forms of abstraction and simplification which contribute to the naturalness of explanations that are communicated to the user.
IsiZulu is one of the eleven official languages of South Africa and roughly half the population can speak it. It is the first (home) language for over 10 million people in South Africa. Only a few computational resources exist for isiZulu and its related Nguni languages, yet the imperative for tool development exists. We focus on natural language generation, and the grammar options and preferences in particular, which will inform verbalization of knowledge representation languages and could contribute to machine translation. The verbalization pattern specification shows that the grammar rules are elaborate and there are several options of which one may have preference. We devised verbalization patterns for subsumption, basic disjointness, existential and universal quantification, and conjunction. This was evaluated in a survey among linguists and non-linguists. Some differences between linguists and non-linguists can be observed, with the former much more in agreement, and preferences depend on the overall structure of the sentence, such as singular for subsumption and plural in other cases.
This paper shows how Long Short-term Memory recurrent neural networks can be used to generate complex sequences with long-range structure, simply by predicting one data point at a time. The approach is demonstrated for text (where the data are discrete) and online handwriting (where the data are real-valued). It is then extended to handwriting synthesis by allowing the network to condition its predictions on a text sequence. The resulting system is able to generate highly realistic cursive handwriting in a wide variety of styles.
A graphical language addresses the need to communicate medical information in a synthetic way. Medical concepts are expressed by icons conveying fast visual information about patients' current state or about the known effects of drugs. In order to increase the visual language's acceptance and usability, a natural language generation interface is currently developed. In this context, this paper describes the use of an informatics method ---graph transformation--- to prepare data consisting of concepts in an OWL-DL ontology for use in a natural language generation component. The OWL concept may be considered as a star-shaped graph with a central node. The method transforms it into a graph representing the deep semantic structure of a natural language phrase. This work may be of future use in other contexts where ontology concepts have to be mapped to half-formalized natural language expressions.
We present in this paper an application which automatically generates textual short-term weather forecasts for every municipality in Galicia (NW Spain), using the real data provided by the Galician Meteorology Agency (MeteoGalicia). This solution combines in an innovative way computing with perceptions techniques and strategies for linguistic description of data together with a natural language generation (NLG) system. The application, named GALiWeather, extracts relevant information from weather forecast input data and encodes it into intermediate descriptions using linguistic variables and temporal references. These descriptions are later translated into natural language texts by the natural language generation system. The obtained forecast results have been thoroughly validated by an expert meteorologist from MeteoGalicia using a quality assessment methodology which covers two key dimensions of a text: the accuracy of its content and the correctness of its form. Following this validation GALiWeather will be released as a real service offering custom forecasts for a wide public.
In this work we present an end-to-end system for text spotting -- localising and recognising text in natural scene images -- and text based image retrieval. This system is based on a region proposal mechanism for detection and deep convolutional neural networks for recognition. Our pipeline uses a novel combination of complementary proposal generation techniques to ensure high recall, and a fast subsequent filtering stage for improving precision. For the recognition and ranking of proposals, we train very large convolutional neural networks to perform word recognition on the whole proposal region at the same time, departing from the character classifier based systems of the past. These networks are trained solely on data produced by a synthetic text generation engine, requiring no human labelled data. Analysing the stages of our pipeline, we show state-of-the-art performance throughout. We perform rigorous experiments across a number of standard end-to-end text spotting benchmarks and text-based image retrieval datasets, showing a large improvement over all previous methods. Finally, we demonstrate a real-world application of our text spotting system to allow thousands of hours of news footage to be instantly searchable via a text query.
In this work we present a framework for the recognition of natural scene text. Our framework does not require any human-labelled data, and performs word recognition on the whole image holistically, departing from the character based recognition systems of the past. The deep neural network models at the centre of this framework are trained solely on data produced by a synthetic text generation engine -- synthetic data that is highly realistic and sufficient to replace real data, giving us infinite amounts of training data. This excess of data exposes new possibilities for word recognition models, and here we consider three models, each one "reading" words in a different way: via 90k-way dictionary encoding, character sequence encoding, and bag-of-N-grams encoding. In the scenarios of language based and completely unconstrained text recognition we greatly improve upon state-of-the-art performance on standard datasets, using our fast, simple machinery and requiring zero data-acquisition costs.
Zipf's law is just one out of many universal laws proposed to describe statistical regularities in language. Here we review and critically discuss how these laws can be statistically interpreted, fitted, and tested (falsified). The modern availability of large databases of written text allows for tests with an unprecedent statistical accuracy and also a characterization of the fluctuations around the typical behavior. We find that fluctuations are usually much larger than expected based on simplifying statistical assumptions (e.g., independence and lack of correlations between observations).These simplifications appear also in usual statistical tests so that the large fluctuations can be erroneously interpreted as a falsification of the law. Instead, here we argue that linguistic laws are only meaningful (falsifiable) if accompanied by a model for which the fluctuations can be computed (e.g., a generative model of the text). The large fluctuations we report show that the constraints imposed by linguistic laws on the creativity process of text generation are not as tight as one could expect.
This paper proposes a novel framework for generating lingual descriptions of indoor scenes. Whereas substantial efforts have been made to tackle this problem, previous approaches focusing primarily on generating a single sentence for each image, which is not sufficient for describing complex scenes. We attempt to go beyond this, by generating coherent descriptions with multiple sentences. Our approach is distinguished from conventional ones in several aspects: (1) a 3D visual parsing system that jointly infers objects, attributes, and relations; (2) a generative grammar learned automatically from training text; and (3) a text generation algorithm that takes into account the coherence among sentences. Experiments on the augmented NYU-v2 dataset show that our framework can generate natural descriptions with substantially higher ROGUE scores compared to those produced by the baseline.
Automatically describing the content of an image is a fundamental problem in artificial intelligence that connects computer vision and natural language processing. In this paper, we present a generative model based on a deep recurrent architecture that combines recent advances in computer vision and machine translation and that can be used to generate natural sentences describing an image. The model is trained to maximize the likelihood of the target description sentence given the training image. Experiments on several datasets show the accuracy of the model and the fluency of the language it learns solely from image descriptions. Our model is often quite accurate, which we verify both qualitatively and quantitatively. For instance, while the current state-of-the-art BLEU-1 score (the higher the better) on the Pascal dataset is 25, our approach yields 59, to be compared to human performance around 69. We also show BLEU-1 score improvements on Flickr30k, from 56 to 66, and on SBU, from 19 to 28. Lastly, on the newly released COCO dataset, we achieve a BLEU-4 of 27.7, which is the current state-of-the-art.
We propose a novel convolutional architecture, named $gen$CNN, for word sequence prediction. Different from previous work on neural network-based language modeling and generation (e.g., RNN or LSTM), we choose not to greedily summarize the history of words as a fixed length vector. Instead, we use a convolutional neural network to predict the next word with the history of words of variable length. Also different from the existing feedforward networks for language modeling, our model can effectively fuse the local correlation and global correlation in the word sequence, with a convolution-gating strategy specifically designed for the task. We argue that our model can give adequate representation of the history, and therefore can naturally exploit both the short and long range dependencies. Our model is fast, easy to train, and readily parallelized. Our extensive experiments on text generation and $n$-best re-ranking in machine translation show that $gen$CNN outperforms the state-of-the-arts with big margins.
Solving the visual symbol grounding problem has long been a goal of artificial intelligence. The field appears to be advancing closer to this goal with recent breakthroughs in deep learning for natural language grounding in static images. In this paper, we propose to translate videos directly to sentences using a unified deep neural network with both convolutional and recurrent structure. Described video datasets are scarce, and most existing methods have been applied to toy domains with a small vocabulary of possible words. By transferring knowledge from 1.2M+ images with category labels and 100,000+ images with captions, our method is able to create sentence descriptions of open-domain videos with large vocabularies. We compare our approach with recent work using language generation metrics, subject, verb, and object prediction accuracy, and a human evaluation.
The formation of sentences is a highly structured and history-dependent process. The probability of using a specific word in a sentence strongly depends on the 'history' of word-usage earlier in that sentence. We study a simple history-dependent model of text generation assuming that the sample-space of word usage reduces along sentence formation, on average. We first show that the model explains the approximate Zipf law found in word frequencies as a direct consequence of sample-space reduction. We then empirically quantify the amount of sample-space reduction in the sentences of ten famous English books, by analysis of corresponding word-transition tables that capture which words can follow any given word in a text. We find a highly nested structure in these transition tables and show that this `nestedness' is tightly related to the power law exponents of the observed word frequency distributions. With the proposed model it is possible to understand that the nestedness of a text can be the origin of the actual scaling exponent, and that deviations from the exact Zipf law can be understood by variations of the degree of nestedness on a book-by-book basis. On a theoretical level we are able to show that in case of weak nesting, Zipf's law breaks down in a fast transition. Unlike previous attempts to understand Zipf's law in language the sample-space reducing model is not based on assumptions of multiplicative, preferential, or self-organised critical mechanisms behind language formation, but simply used the empirically quantifiable parameter 'nestedness' to understand the statistics of word frequencies.
Natural language generation of coherent long texts like paragraphs or longer documents is a challenging problem for recurrent networks models. In this paper, we explore an important step toward this generation task: training an LSTM (Long-short term memory) auto-encoder to preserve and reconstruct multi-sentence paragraphs. We introduce an LSTM model that hierarchically builds an embedding for a paragraph from embeddings for sentences and words, then decodes this embedding to reconstruct the original paragraph. We evaluate the reconstructed paragraph using standard metrics like ROUGE and Entity Grid, showing that neural models are able to encode texts in a way that preserve syntactic, semantic, and discourse coherence. While only a first step toward generating coherent text units from neural models, our work has the potential to significantly impact natural language generation and summarization\footnote{Code for the three models described in this paper can be found at www.stanford.edu/~jiweil/ .
We propose a simple, scalable, fully generative model for transition-based dependency parsing with high accuracy. The model, parameterized by Hierarchical Pitman-Yor Processes, overcomes the limitations of previous generative models by allowing fast and accurate inference. We propose an efficient decoding algorithm based on particle filtering that can adapt the beam size to the uncertainty in the model while jointly predicting POS tags and parse trees. The UAS of the parser is on par with that of a greedy discriminative baseline. As a language model, it obtains better perplexity than a n-gram model by performing semi-supervised learning over a large unlabelled corpus. We show that the model is able to generate locally and syntactically coherent sentences, opening the door to further applications in language generation.
We present a general theory and corresponding declarative model for the embodied grounding and natural language based analytical summarisation of dynamic visuo-spatial imagery. The declarative model ---ecompassing spatio-linguistic abstractions, image schemas, and a spatio-temporal feature based language generator--- is modularly implemented within Constraint Logic Programming (CLP). The implemented model is such that primitives of the theory, e.g., pertaining to space and motion, image schemata, are available as first-class objects with `deep semantics' suited for inference and query. We demonstrate the model with select examples broadly motivated by areas such as film, design, geography, smart environments where analytical natural language based externalisations of the moving image are central from the viewpoint of human interaction, evidence-based qualitative analysis, and sensemaking. Keywords: moving image, visual semantics and embodiment, visuo-spatial cognition and computation, cognitive vision, computational models of narrative, declarative spatial reasoning