scispace - formally typeset
Search or ask a question
Topic

Natural language understanding

About: Natural language understanding is a research topic. Over the lifetime, 2577 publications have been published within this topic receiving 51905 citations.


Papers
More filters
Proceedings Article
10 Jun 2012
TL;DR: The Winograd Schema Challenge as mentioned in this paper is an alternative to the Turing Test that has some conceptual and practical advantages, such as the ability to be easily found using selectional restrictions or statistical techniques over text corpora.
Abstract: In this paper, we present an alternative to the Turing Test that has some conceptual and practical advantages. A Wino-grad schema is a pair of sentences that differ only in one or two words and that contain a referential ambiguity that is resolved in opposite directions in the two sentences. We have compiled a collection of Winograd schemas, designed so that the correct answer is obvious to the human reader, but cannot easily be found using selectional restrictions or statistical techniques over text corpora. A contestant in the Winograd Schema Challenge is presented with a collection of one sentence from each pair, and required to achieve human-level accuracy in choosing the correct disambiguation.

928 citations

01 Jan 1995
TL;DR: How lexical chains can be constructed by means of WordNet, and how they can be applied in one particularlinguistic task: the detection and correction of malapropisms is shown.
Abstract: Natural language utterances are, in general, highlyambiguous, and a unique interpretationcan usuallybe determined only by taking into account the constraining influence of the context in which theutterance occurred. Much of the research in natural language understanding in the last twenty yearscan be thought of as attempts to characterize and represent context and then derive interpretationsthatfit best with that context. Typically, this research was heavy with AI, taking context to be nothing lessthan a complete conceptual understanding of the preceding utterances. This was reasonable, as suchan understanding of a text was often the main task anyway. However, there are many text-processingtasksthatrequireonlya partialunderstandingofthetext, andhencea ‘lighter’representationofcontextis sufficient. In this paper, we examine the idea oflexical chains as such a representation. We showhow they can be constructed by means of WordNet, and how they can be applied in one particularlinguistic task: the detection and correction of malapropisms.A malapropism is the confounding of an intended word with another word of similar sound orsimilar spelling that has a quite different and malapropos meaning, e.g., an ingenuous [for ingenious]machine forpeelingoranges. In thisexample, there isaone-letterdifference betweenthe malapropismand the correct word. Ignorance, or a simple typing mistake, might cause such errors. However, sinceingenuous is a correctly spelled word, traditional spelling checkers cannot detect this kind of mistake.In section 4, we will propose an algorithm for detecting and correcting malapropisms that is based onthe construction of lexical chains.

915 citations

Proceedings Article
07 Aug 2011
TL;DR: A learning process based on an innovative neural network architecture designed to embed any of these symbolic representations into a more flexible continuous vector space in which the original knowledge is kept and enhanced would allow data from any KB to be easily used in recent machine learning methods for prediction and information retrieval.
Abstract: Many Knowledge Bases (KBs) are now readily available and encompass colossal quantities of information thanks to either a long-term funding effort (e.g. WordNet, OpenCyc) or a collaborative process (e.g. Freebase, DBpedia). However, each of them is based on a different rigid symbolic framework which makes it hard to use their data in other systems. It is unfortunate because such rich structured knowledge might lead to a huge leap forward in many other areas of AI like natural language processing (word-sense disambiguation, natural language understanding, ...), vision (scene classification, image semantic annotation, ...) or collaborative filtering. In this paper, we present a learning process based on an innovative neural network architecture designed to embed any of these symbolic representations into a more flexible continuous vector space in which the original knowledge is kept and enhanced. These learnt embeddings would allow data from any KB to be easily used in recent machine learning methods for prediction and information retrieval. We illustrate our method on WordNet and Freebase and also present a way to adapt it to knowledge extraction from raw text.

909 citations

Posted Content
TL;DR: AllenNLP is designed to support researchers who want to build novel language understanding models quickly and easily and provides a flexible data API that handles intelligent batching and padding, and a modular and extensible experiment framework that makes doing good science easy.
Abstract: This paper describes AllenNLP, a platform for research on deep learning methods in natural language understanding. AllenNLP is designed to support researchers who want to build novel language understanding models quickly and easily. It is built on top of PyTorch, allowing for dynamic computation graphs, and provides (1) a flexible data API that handles intelligent batching and padding, (2) high-level abstractions for common operations in working with text, and (3) a modular and extensible experiment framework that makes doing good science easy. It also includes reference implementations of high quality approaches for both core semantic problems (e.g. semantic role labeling (Palmer et al., 2005)) and language understanding applications (e.g. machine comprehension (Rajpurkar et al., 2016)). AllenNLP is an ongoing open-source effort maintained by engineers and researchers at the Allen Institute for Artificial Intelligence.

767 citations


Network Information
Related Topics (5)
Natural language
31.1K papers, 806.8K citations
88% related
Recurrent neural network
29.2K papers, 890K citations
81% related
Ontology (information science)
57K papers, 869.1K citations
81% related
Graph (abstract data type)
69.9K papers, 1.2M citations
79% related
Deep learning
79.8K papers, 2.1M citations
79% related
Performance
Metrics
No. of papers in the topic in previous years
YearPapers
202395
2022189
2021319
2020357
2019274
2018161