The theoretical code-switching (CS) literature provides numerous pointwise investigations that aim to explain patterns in CS, i.e. why bilinguals switch language in certain positions in a sentence more often than in others. A resulting consensus is that CS can be explained by the syntax of the contributing languages. There is however no large-scale, multi-language, cross-phenomena experiment that tests this claim. When designing such an experiment, we need to make sure that the system that is predicting where bilinguals tend to switch has access only to syntactic information. We provide such an experiment here. Results show that syntax alone is sufficient for an automatic system to distinguish between sentences in minimal pairs of CS, to the same degree as bilingual humans. Furthermore, the learnt syntactic patterns generalise well to unseen language pairs.

What would happen if temperatures were subdued and result in a cool summer? One can easily imagine that air conditioner, ice cream or beer sales would be suppressed as a result of this. Less obvious is that agricultural shipments might be delayed, or that sound proofing material sales might decrease. The ability to extract such causal knowledge is important, but it is also important to distinguish between cause-effect pairs that are known and those that are likely to be unknown, or rare. Therefore, in this paper, we propose a method for extracting rare causal knowledge from Japanese financial statement summaries produced by companies. Our method consists of three steps. First, it extracts sentences that include causal knowledge from the summaries using a machine learning method based on an extended language ontology. Second, it obtains causal knowledge from the extracted sentences using syntactic patterns. Finally, it extracts the rarest causal knowledge from the knowledge it has obtained.

It is an established assumption that pattern-based models are good at precision, while learning based models are better at recall. But is that really the case? I argue that there are two kinds of recall: d-recall, reflecting diversity, and e-recall, reflecting exhaustiveness. I demonstrate through experiments that while neural methods are indeed significantly better at d-recall, it is sometimes the case that pattern-based methods are still substantially better at e-recall. Ideal methods should aim for both kinds, and this ideal should in turn be reflected in our evaluations.

Internet users have been exposing an increasing amount of Personally Identifiable Information (PII) on social media. Such exposed PII can cause severe losses to the users, and informing users of their PII exposure is crucial to raise their privacy awareness and encourage them to take protective measures. To this end, advanced automatic techniques are needed. While Information Extraction (IE) techniques can be used to extract the PII automatically, Deep Learning (DL)-based IE models alleviate the need for feature engineering and further improve the efficiency. However, DL-based IE models often require large-scale labeled data for training, but PII-labeled social media posts are difficult to obtain due to privacy concerns. Also, these models rely heavily on pre-trained word embeddings, while PII in social media often varies in forms and thus has no fixed representations in pre-trained word embeddings. In this study, we propose the Deep Transfer Learning for PII Extraction (DTL-PIIE) framework to address these two limitations. DTL-PIIE transfers knowledge learned from publicly available PII data to social media to address the problem of rare PII-labeled data. Moreover, our framework leverages Graph Convolutional Networks (GCNs) to incorporate syntactic patterns to guide PIIE without relying on pre-trained word embeddings. Evaluation against benchmark IE models indicates that our approach outperforms state-of-the-art DL-based IE models. Our framework can facilitate various applications, such as PII misuse prediction and privacy risk assessment, protecting the privacy of internet users.

The vehicle to represent Knowledge Organization Systems (KOSs) in the environment of the Semantic Web and linked data is the Simple Knowledge Organization System (SKOS). SKOS provides a way to assign a URI to each concept, and this URI functions as a surrogate for the concept. This fact makes of main concern the need to clarify the URIs' ontological meaning. The aim of this study is to investigate the relation between the ontological substance of KOS concepts and concepts revealed through the grammatical and syntactic formalisms of natural language. For this purpose, we examined the dividableness of concepts in specific KOSs (i.e. a thesaurus, a subject headings system and a classification scheme) by applying Natural Language Processing (NLP) techniques (i.e. morphosyntactic analysis) to the lexical representations (i.e. RDF literals) of SKOS concepts. The results of the comparative analysis reveal that, despite the use of multi-word units, thesauri tend to represent concepts in a way that can hardly be further divided conceptually, while Subject Headings and Classification Schemes - to a certain extent - comprise terms that can be decomposed into more conceptual constituents. Consequently, SKOS concepts deriving from thesauri are more likely to represent atomic conceptual units and thus be more appropriate tools for inference and reasoning. Since identifiers represent the meaning of a concept, complex concepts are neither the most appropriate nor the most efficient way of modelling a KOS for the Semantic Web.

Computer scientists at MIT and Israel's Technion have discovered an unexpected source of information about the world's languages: the habits of native speakers of those languages when writing in English. The work could enable computers chewing through relatively accessible documents to approximate data that might take trained linguists months in the field to collect. But that data could in turn lead to better computational tools. "These [linguistic] features that our system is learning are of course, on one hand, of nice theoretical interest for linguists," says Boris Katz, a principal research scientist at MIT's Computer Science and Artificial Intelligence Laboratory and one of the leaders of the new work. "But on the other, they're beginning to be used more and more often in applications. Everybody's very interested in building computational tools for world languages, but in order to build them, you need these features. So we may be able to do much more than just learn linguistic ...

We address the problem of generating all possible palindromes from a corpus of Ngrams. Palindromes are texts that read the same both ways. Short palindromes ("race car") usually carry precise, significant meanings. Long palindromes are often less meaningful, but even harder to generate. The palindrome generation problem has never been addressed, to our knowledge, from a strictly combinatorial point of view. The main difficulty is that generating palindromes require the simultaneous consideration of two inter-related levels in a sequence: the "character" and the "word" levels. Although the problem seems very combinatorial, we propose an elegant yet non-trivial graph structure that can be used to generate all possible palindromes from a given corpus of Ngrams, with a linear complexity. We illustrate our approach with short and long palindromes obtained from the Google Ngram corpus. We show how we can control the semantics, to some extent, by using arbitrary text corpora to bias the probabilities of certain sets of words. More generally this work addresses the issue of modelling human virtuosity from a combinatorial viewpoint, as a means to understand human creativity.

This paper discusses the implementation and evaluation of automatically generated cloze questions in the style of the definitions found in Collins COBUILD English language learner’s dictionary. The definitions and the cloze questions are used in an automated reading tutor to help second and third grade students learn new vocabulary. A parser provides syntactic phrase structure trees for the definitions. With these parse trees as input, a pattern matching program uses a set of syntactic patterns to extract the phrases that make up the cloze question answers and distracters.

At a computational level, language processing tasks are traditionally processed in a language-only space/context, isolated from perception and action. However, at a cognitive level, language processing has been shown experimentally to be embodied, i.e. to inform and be informed by perception and action. In this paper, we argue that embodied cognition dictates the development of a new generation of language processing tools that bridge the gap between the symbolic and the sensorimotor representation spaces. We describe that tasks and challenges such tools need to address and provide an overview of the first such suite of processing tools developed in the framework of the POETICON project.

Extracting the relations that exist between two entities is an important step in numerousWeb-related tasks such as information extraction.A supervised relation extraction system that is trained to extract a particular relation type might not accurately extract a new type of a relation for which it has not been trained.However, it is costly to create training data manually for every new relation type that one might want to extract.We propose a method to adapt an existing relation extraction system to extractnew relation types with minimum supervision. Our proposed method comprises two stages: learning a lower-dimensional projection between different relations, and learning a relational classifier for the target relation type with instance sampling. We evaluate the proposed method using a dataset that contains 2000 instances for 20 different relation types. Our experimental results show that the proposed method achieves a statistically significant macro-average F-score of 62.77. Moreover, the proposed method outperforms numerous baselines and a previously proposed weakly-supervised relation extraction method.