The FrameNet database comprises an English lexicon, organized in terms of semantic frames. Frames describe situations or entities, along with their participants and props, termed frame elements. The frames are organized in an ontology-like network. For the lexical units, corpus annotations illustrate which frame elements are typically realized, and how they behave syntactically. Texts where all content words are annotated with FrameNet information offer a detailed, structured semantic representation with a variety of uses in Natural Language Processing applications, in particular in retrieving and meaningfully organizing texts written by humans, or in making human–computer interaction more natural. Also, the FrameNet English lexicon can be replaced by lexical data from other languages, while maintaining frame information, so the model is attractive for cross-lingual resources and applications. Manual annotation produced by FrameNet and similar projects for other languages is used to train automatic frame semantic annotation systems, which add rich semantic information to any type of text, and are important components for more sophisticated semantic processing applications.

In this article, we describe AIMS (Assisted Indexing at Mississippi State), a system intended to aid human document analysts in the assignment of indexes to physical chemistry journal articles. The two major components of AIMS are a natural language processing (NLP) component and an index generation (IG) component. We provide an overview of what each of these components does and how it works. We also present the results of a recent evaluation of our system in terms of recall and precision. The recall rate is the proportion of the ‘correct’ indexes (i.e. those produced by human document analysts) generated by AIMS. The precision rate is the proportion of the generated indexes that is correct. Finally, we describe some of the future work planned for this project.

In this paper, the August spoken dialogue system is described. This experimental Swedish dialogue system, which featured an animated talking agent, was exposed to the general public during a trial period of six months. The construction of the system was partly motivated by the need to collect genuine speech data from people with little or no previous experience of spoken dialogue systems. A corpus of more than 10,000 utterances of spontaneous computer- directed speech was collected and empirical linguistic analyses were carried out. Acoustical, lexical and syntactical aspects of this data were examined. In particular, user behavior and user adaptation during error resolution were emphasized. Repetitive sequences in the database were analyzed in detail. Results suggest that computer-directed speech during error resolution is increased in duration, hyperarticulated and contains inserted pauses. Design decisions which may have influenced how the users behaved when they interacted with August are discussed and implications for the development of future systems are outlined.

This paper presents an investigation of part of speech (POS) tagging for Arabic as it occurs naturally, i.e. unvocalized text (without diacritics). We also do not assume any prior tokenization, although this was used previously as a basis for POS tagging. Arabic is a morphologically complex language, i.e. there is a high number of inflections per word; and the tagset is larger than the typical tagset for English. Both factors, the second one being partly dependent on the first, increase the number of word/tag combinations, for which the POS tagger needs to find estimates, and thus they contribute to data sparseness. We present a novel approach to Arabic POS tagging that does not require any pre-processing, such as segmentation or tokenization: whole word tagging. In this approach, the complete word is assigned a complex POS tag, which includes morphological information. A competing approach investigates the effect of segmentation and vocalization on POS tagging to alleviate data sparseness and ambiguity. In the segmentation-based approach, we first automatically segment words and then POS tags the segments. The complex tagset encompasses 993 POS tags, whereas the segment-based tagset encompasses only 139 tags. However, segments are also more ambiguous, thus there are more possible combinations of segment tags. In realistic situations, in which we have no information about segmentation or vocalization, whole word tagging reaches the highest accuracy of 94.74%. If gold standard segmentation or vocalization is available, including this information improves POS tagging accuracy. However, while our automatic segmentation and vocalization modules reach state-of-the-art performance, their performance is not reliable enough for POS tagging and actually impairs POS tagging performance. Finally, we investigate whether a reduction of the complex tagset to the Extra-Reduced Tagset as suggested by Habash and Rambow (Habash, N., and Rambow, O. 2005. Arabic tokenization, part-of-speech tagging and morphological disambiguation in one fell swoop. In Proceedings of the 43rd Annual Meeting of the Association for Computational Linguistics (ACL), Ann Arbor, MI, USA, pp. 573–80) will alleviate the data sparseness problem. While the POS tagging accuracy increases due to the smaller tagset, a closer look shows that using a complex tagset for POS tagging and then converting the resulting annotation to the smaller tagset results in a higher accuracy than tagging using the smaller tagset directly.

This paper presents modifications to a standard probabilistic context-free grammar that enable a predictive parser to avoid garden pathing without resorting to any ad-hoc heuristic repair. The resulting parser is shown to apply efficiently to both newspaper text and telephone conversations with complete coverage and excellent accuracy. The distribution over trees is peaked enough to allow the parser to find parses efficiently, even with the much larger search space resulting from overgeneration. Empirical results are provided for both Wall St. Journal and Switchboard test corpora.

Hyphenation is the task of identifying potential hyphenation points inwords. In this paper, three finite-state hyphenation methods for Dutch are presented and compared in terms of accuracy and size of the resulting automata.

This paper describes the approach to knowledge representation taken in the LaSIE Information Extraction (IE) system. Unlike many IE systems that skim texts and use large collections of shallow, domain-specific patterns and heuristics to fill in templates, LaSIE attempts a fuller text analysis, first translating individual sentences to a quasi-logical form, and then constructing a weak discourse model of the entire text from which template fills are finally derived. Underpinning the system is a general ‘world model’, represented as a semantic net, which is extended during the processing of a text by adding the classes and instances described in that text. In the paper we describe the system's knowledge representation formalisms, their use in the IE task, and how the knowledge represented in them is acquired, including experiments to extend the system's coverage using the WordNet general purpose semantic network. Preliminary evaluations of our approach, through the Sixth DARPA Message Understanding Conference, indicate comparable performance to shallower approaches. However, we believe its generality and extensibility offer a route towards the higher precision that is required of IE systems if they are to become genuinely usable technologies.

We describe and argue for a strategy of performance profiling and comparison in the engineering of parsing systems for wide-coverage linguistic grammars. A performance profile is a precise, rich and structured snapshot of system (and grammar) behaviour at a given development point. The aim is to characterize system performance at a very detailed technical level, but at the same time to abstract away from idiosyncracies of particular processors. Profiles are obtained with minimal effort by applying a specialized profiling tool to a set of structured reference data (taken from both existing test suites and corpora), in conjunction with a uniform format for test data and processing results. The resulting profiles can be analyzed and visualized at various levels of granularity in order to highlight different aspects of system performance, thus providing a solid empirical basis for system refinement and optimization. Since profiles are stored in a database, comparison with earlier versions, different parameter settings, or other processing systems is straightforward. We apply several salient performance metrics in a contrastive discussion of various (one-pass, bottom-up, chart-based) parsing strategies (viz. passive vs. active and uni- vs. bidirectional approaches). Based on insights gained from detailed performance profiles, we outline and evaluate a novel ‘hyper-active’ parsing strategy. We also present preliminary profiles for techniques for ‘packing’ of local ambiguities with respect to (partial) subsumption of feature structures.

To work effectively in information-rich environments, knowledge workers must be able to distil the most appropriate information from the deluge of information available to them. This is difficult to do manually. Natural language engineers can support these workers by developing information delivery tools, but because of the wide variety of contexts in which information is acquired and delivered, these tools have tended to be domain-specific, ad hoc solutions that are hard to generalise. This paper discusses Myriad, a platform that generalises the integration of sets of resources to a variety of information delivery contexts. Myriad provides resources from natural language generation for discourse planning as well as a service-based architecture for data access. The nature of Myriad's resources is driven by engineering concerns. It focuses on resources that reason about and generate from coarse-grained units of information, likely to be provided by existing information sources, and it supports the integration of pipe-lined planning and template mechanisms. The platform is illustrated in the context of three information delivery applications and is evaluated with respect to its utility.

In this paper we discuss the task of dialogue act recognition as a part of interpreting user utterances in context. To deal with the uncertainty that is inherent in natural language processing in general and dialogue act recognition in particular we use machine learning techniques to train classifiers from corpus data. These classifiers make use of both lexical features of the (Dutch) keyboard-typed utterances in the corpus used, and context features in the form of dialogue acts of previous utterances. In particular, we consider probabilistic models in the form of Bayesian networks to be proposed as a more general framework for dealing with uncertainty in the dialogue modelling process.

Richly annotated dialogue corpora are essential for new research directions in statistical learning approaches to dialogue management, context-sensitive interpretation, and context-sensitive speech recognition. In particular, large dialogue corpora annotated with contextual information and speech acts are urgently required. We explore how existing dialogue corpora (usually consisting of utterance transcriptions) can be automatically processed to yield new corpora where dialogue context and speech acts are accurately represented. We present a conceptual and computational framework for generating such corpora. As an example, we present and evaluate an automatic annotation system which builds ‘Information State Update’ (ISU) representations of dialogue context for the Communicator (2000 and 2001) corpora of human–machine dialogues (2,331 dialogues). The purposes of this annotation are to generate corpora for reinforcement learning of dialogue policies, for building user simulations, for evaluating different dialogue strategies against a baseline, and for training models for context-dependent interpretation and speech recognition. The automatic annotation system parses system and user utterances into speech acts and builds up sequences of dialogue context representations using an ISU dialogue manager. We present the architecture of the automatic annotation system and a detailed example to illustrate how the system components interact to produce the annotations. We also evaluate the annotations, with respect to the task completion metrics of the original corpus and in comparison to hand-annotated data and annotations produced by a baseline automatic system. The automatic annotations perform well and largely outperform the baseline automatic annotations in all measures. The resulting annotated corpus has been used to train high-quality user simulations and to learn successful dialogue strategies. The final corpus will be made publicly available.

A new framework for improving event detection is proposed that employs joint information in news media content and social networks, such as Twitter, to leverage detailed coverage of news media and the timeliness of social media. Specifically, a short text clustering method is employed to detect events from tweets, then the language model representations of the detected events are expanded using another set of events obtained from news articles published simultaneously. The expanded representations of events are employed as a new initialization of the clustering method to run another iteration and consequently enhance the event detection results. The proposed framework is evaluated using two datasets: a tweet dataset with event labels and a news dataset containing news articles published during the same time interval as the tweets. Experimental results show that the proposed framework improves the event detection results in terms of F1 measure compared to the results obtained from tweets only.

We present the novel task of understanding multi-sentence entity-seeking questions (MSEQs), that is, the questions that may be expressed in multiple sentences, and that expect one or more entities as an answer. We formulate the problem of understanding MSEQs as a semantic labeling task over an open representation that makes minimal assumptions about schema or ontology-specific semantic vocabulary. At the core of our model, we use a BiLSTM (bidirectional LSTM) conditional random field (CRF), and to overcome the challenges of operating with low training data, we supplement it by using BERT embeddings, hand-designed features, as well as hard and soft constraints spanning multiple sentences. We find that this results in a 12–15 points gain over a vanilla BiLSTM CRF. We demonstrate the strengths of our work using the novel task of answering real-world entity-seeking questions from the tourism domain. The use of our labels helps answer 36% more questions with 35% more (relative) accuracy as compared to baselines. We also demonstrate how our framework can rapidly enable the parsing of MSEQs in an entirely new domain with small amounts of training data and little change in the semantic representation.

Poetry generation is a specific kind of natural language generation where several sources of knowledge are typically exploited to handle features on different levels, such as syntax, semantics, form or aesthetics. But although this task has been addressed by several researchers, and targeted different languages, all known systems have focused on a limited purpose and a single language. This article describes the effort of adapting the same architecture to generate poetry in three different languages – Portuguese, Spanish and English. An existing architecture is first described and complemented with the adaptations required for each language, including the linguistic resources used for handling morphology, syntax, semantics and metric scansion. An automatic evaluation was designed in such a way that it would be applicable to the target languages. It covered three relevant aspects of the generated poems, namely: the presence of poetic features, the variation of the linguistic structure and the semantic connection to a given topic. The automatic measures applied for the second and third aspect can be seen as novel in the evaluation of poetry. Overall, poems were successfully generated in the three languages addressed. Despite minor differences in different languages or seed words, poems revealed to have a regular metre, frequent rhymes, to exhibit an interesting degree of variation, and to be semantically-associated with the initially given seeds.

This paper presents a system developed for detecting sexual predators in online chat conversations using a two-stage classification and behavioral features. A sexual predator is defined as a person who tries to obtain sexual favors in a predatory manner, usually with underage people. The proposed approach uses several text categorization methods and empirical behavioral features developed especially for the task at hand. After investigating various approaches for solving the sexual predator identification problem, we have found that a two-stage classifier achieves the best results. In the first stage, we employ a Support Vector Machine classifier to distinguish conversations having suspicious content from safe online discussions. This is useful as most chat conversations in real life do not contain a sexual predator, therefore it can be viewed as a filtering phase that enables the actual detection of predators to be done only for suspicious chats that contain a sexual predator with a very high degree. In the second stage, we detect which of the users in a suspicious discussion is an actual predator using a Random Forest classifier. The system was tested on the corpus provided by the PAN 2012 workshop organizers and the results are encouraging because, as far as we know, our solution outperforms all previous approaches developed for solving this task.

We describe an interactive question answering system, HITIQA, which helps users find answers to complex analytical problems. Such problems often necessitate the user to submit not one but an entire series of questions, both simple and complex, and then to negotiate the final content and form of the answer. HITIQA advances research in human–computer dialogue by enabling topical, mixed initiative interaction over unstructured data. HITIQA uses the process of text framing to bring a level of semantic representation to open-domain data in order to facilitate meaningful dialogue with the user. In this paper we give an overview of HITIQA's design and explain the workings of its main components with particular attention given to its dialogue capabilities. We also present results of end-to-end system evaluations that demonstrate the effectiveness of the system as a whole, as well as contributions of the individual components and specifically the benefits of our dialogue-based approach. While our research continues, a number of HITIQA prototypes have recently been deployed at various government agencies where they are being tested under real operational conditions.

We present a method for mining the web for text entered on mobile devices. Using searching, crawling, and parsing techniques, we locate text that can be reliably identified as originating from 300 mobile devices. This includes 341,000 sentences written on iPhones alone. Our data enables a richer understanding of how users type “in the wild” on their mobile devices. We compare text and error characteristics of different device types, such as touchscreen phones, phones with physical keyboards, and tablet computers. Using our mined data, we train language models and evaluate these models on mobile test data. A mixture model trained on our mined data, Twitter, blog, and forum data predicts mobile text better than baseline models. Using phone and smartwatch typing data from 135 users, we demonstrate our models improve the recognition accuracy and word predictions of a state-of-the-art touchscreen virtual keyboard decoder. Finally, we make our language models and mined dataset available to other researchers.

With NLP services now widely available via cloud APIs, tasks like named entity recognition and sentiment analysis are virtually commodities. We look at what's on offer, and make some suggestions for how to get rich.

Usage of large language models and chat bots will almost surely continue to grow, since they are so easy to use, and so (incredibly) credible. I would be more comfortable with this reality if we encouraged more evaluations with humans-in-the-loop to come up with a better characterization of when the machine can be trusted and when humans should intervene. This article will describe a homework assignment, where I asked my students to use tools such as chat bots and web search to write a number of essays. Even after considerable discussion in class on hallucinations, many of the essays were full of misinformation that should have been fact-checked. Apparently, it is easier to believe ChatGPT than to be skeptical. Fact-checking and web search are too much trouble.

We investigate the use of different machine learning methods in combination with feature selection techniques to explore human multimodal dialogue strategies and the use of those strategies for automated dialogue systems. We learn policies from data collected in a Wizard-of-Oz study where different human ‘wizards’ decide whether to ask a clarification request in a multimodal manner or else to use speech alone. We first describe the data collection, the coding scheme and annotated corpus, and the validation of the multimodal annotations. We then show that there is a uniform multimodal dialogue strategy across wizards, which is based on multiple features in the dialogue context. These are generic features, available at runtime, which can be implemented in dialogue systems. Our prediction models (for human wizard behaviour) achieve a weighted f-score of 88.6 per cent (which is a 25.6 per cent improvement over the majority baseline). We interpret and discuss the learned strategy. We conclude that human wizard behaviour is not optimal for automatic dialogue systems, and argue for the use of automatic optimization methods, such as Reinforcement Learning. Throughout the investigation we also discuss the issues arising from using small initial Wizard-of-Oz data sets, and we show that feature engineering is an essential step when learning dialogue strategies from such limited data.