Search

In this paper, the relation between the FAIR (Findable, Accessible, Interoperable, Reusable) ontologies and accountability and transparency of ontology-based AI systems is analysed. Also, governance-related gaps in ontology quality evaluation metrics were identified by examining their relation with FAIR principles and FAcct (Fairness, Accountability, Transparency) governance aspects. A simple SKOS vocabulary, titled "Hospital Adverse Incidents Classification Scheme" (HAICS) has been used as a use case for this study. Theoretically, we found that there is a straight relation between FAIR principles and FAccT AI, which means that FAIR ontologies enhance transparency and accountability in ontology-based AI systems. We suggest that "FAIRness" should be assessed as one of the ontology quality evaluation aspects.

Query auto-completion helps users to formulate their information needs by providing suggestion lists at every typed key. This task is commonly addressed by exploiting query logs and the approaches proposed in the literature fit well in web scale scenarios, where usually huge amounts of past user queries can be analyzed to provide reliable suggestions. However, when query logs are not available, e.g. in enterprise or desktop search engines, these methods are not applicable at all. To face these challenging scenarios, we present a novel corpus-based approach which exploits the textual content of an indexed document collection in order to dynamically generate query completions. Our method extracts informative text fragments from the corpus and it combines them using a probabilistic graphical model in order to capture the relationships between the extracted concepts. Using this approach, it is possible to automatically complete partial queries with significant suggestions related to the keywords already entered by the user without requiring the analysis of the past queries. We evaluate our system through a user study on two different real-world document collections. The experiments show that our method is able to provide meaningful completions outperforming the state-of-the art approach.

In this paper, we describe the creation of a diachronic corpus for Italian by exploiting the digital archive of the newspaper “L’Unit`a”. We automatically clean and annotate the corpus with PoStags, lemmas, named entities and syntactic dependencies. Moreover, we compute frequency-based time series for tokens, lemmas and entities. We show some interesting corpus statistics taking into account the temporal dimension and describe some examples of usage of time series.

This paper describes the system proposed by the Random team for SemEval-2020 Task 1: Unsupervised Lexical Semantic Change Detection. We focus our approach on the detection problem. Given the semantics of words captured by temporal word embeddings in different time periods, we investigate the use of unsupervised methods to detect when the target word has gained or lost senses. To this end, we define a new algorithm based on Gaussian Mixture Models to cluster the target similarities computed over the two periods. We compare the proposed approach with a number of similarity-based thresholds. We found that, although the performance of the detection methods varies across the word embedding algorithms, the combination of Gaussian Mixture with Temporal Referencing resulted in our best system.

In the last few years, the increasing availability of large corpora spanning several time periods has opened new opportunities for the diachronic analysis of language. This type of analysis can bring to the light not only linguistic phenomena related to the shift of word meanings over time, but it can also be used to study the impact that societal and cultural trends have on this language change. This paper introduces a new resource for performing the diachronic analysis of named entities built upon Wikipedia page revisions. This resource enables the analysis over time of changes in the relations between entities (concepts), surface forms (words), and the contexts surrounding entities and surface forms, by analysing the whole history of Wikipedia internal links. We provide some useful use cases that prove the impact of this resource on diachronic studies and delineate some possible future usage. Source code and more details are available here: https://github.com/pippokill/dae

Linking entity mentions in Italian tweets to concepts in a knowledge base is a challenging task, due to the short and noisy nature of these short messages and the lack of specific resources for Italian. This paper proposes an adaptation of a general purpose Named Entity Linking algorithm, which exploits the similarity measure computed over a Distributional Semantic Model, in the context of Italian tweets. In order to evaluate the proposed algorithm, we introduce a new dataset of tweets for entity linking that we have developed specifically for the Italian language.

During the last decade the surge in available data spanning different epochs has inspired a new analysis of cultural, social, and linguistic phenomena from a temporal perspective. This paper describes a method that enables the analysis of the time evolution of the meaning of a word. We propose Temporal Random Indexing (TRI), a method for building WordSpaces that takes into account temporal information. We exploit this methodology in order to build geometrical spaces of word meanings that consider several periods of time. The TRI framework provides all the necessary tools to build WordSpaces over different time periods and perform such temporal linguistic analysis. We propose some examples of usage of our tool by analysing word meanings in two corpora: a collection of Italian books and English scientific papers about computational linguistics. This analysis enables the detection of linguistic events that emerge in specific time intervals and that can be related to social or cultural phenomena.

The growth of the Web is the most influential factor that contributes to the increasing importance of text retrieval and filtering systems. On one hand, the Web is becoming more and more multilingual, and on the other hand users themselves are becoming increasingly polyglot. In this context, platforms for intelligent information access as search engines or recommender systems need to evolve to deal with this increasing amount of multilingual information. This paper proposes a content-based recommender system able to generate cross-lingual recommendations. The idea is to exploit user preferences learned in a given language, to suggest item in another language. The main intuition behind the work is that, differently from keywords which are inherently language dependent, concepts are stable across different languages, allowing to deal with multilingual and cross-lingual scenarios. We propose four knowledge-based strategies to build concept-based representation of items, by relying on the knowledge contained in two knowledge sources, i.e. Wikipedia and BabelNet. We learn user profiles by leveraging the different concept-based representations, in order to define a cross-lingual recommendation process. The empirical evaluation carried out on two state of the art datasets, DBbook and Movielens, shows that concept-based approaches are suitable to provide cross-lingual recommendations, even though there is not a clear advantage of using one of the different proposed representations. However, it emerges that most of the times the approaches based on BabelNet outperform those based on Wikipedia, which clearly shows the advantage of using a native multilingual knowledge source.

This paper describes the effect of introducing embedding-based features in a learning to rank approach to entity relatedness. We define several features that exploit word- and link-embedding approaches by relying on both links and the content that appear in Wikipedia articles. These features are combined with other state-of-the-art relatedness measures by using a learning to rank framework. In the evaluation, we report the performance of each feature individually. Moreover, we investigate the contribution of each feature to the ranking function by analysing the output of a feature selection algorithm. The results of this analysis prove that features based on word and link embeddings are able to increase the performance of the learning to rank algorithm.

Named Entity Linking (NEL) is the task of semantically annotating entity mentions in a portion of text with links to a knowledge base. The automatic annotation, which requires the recognition and disambiguation of the entity mention, usually exploits contextual clues like the context of usage and the coherence with respect to other entities. In Twitter, the limits of 140 characters originates very short and noisy text messages that pose new challenges to the entity linking task. We propose an overview of NEL methods focusing on approaches specifically developed to deal with short messages, like tweets. NEL is a fundamental task for the extraction and annotation of concepts in tweets, which is necessary for making the Twitter’s huge amount of interconnected user-generated contents machine readable and enable the intelligent information access.

This paper introduces a semantic and personalised information retrieval (SEPIR) tool for the public administration of Apulia Region. SEPIR, through semantic search and visualisation tools, enables the analysis of a large amount of unstructured data and the intelligent access to information. At the core of these functionalities is an NLP pipeline responsible for the WordSpace building and the key-phrase extraction. The WordSpace is the key component of the semantic search and personalisation algorithm. Moreover, key-phrases enrich the document representation of the retrieval system and are on the basis of the bubble charts, which provide a quick overview of the main concepts involved in a document collection. We show some of the key features of SEPIR in a use case where the personalisation technique re-ranks the set of relevant documents on the basis of the user's past queries and the visualisation tools provide the users with useful information about the analysed collection.

Super-sense tagging is the task of annotating each word in a text with a super-sense, i.e. a general concept such as animal, food or person, coming from the general semantic taxonomy defined by the WordNet lexicographer classes. Due to the small set of involved concepts, the task is simpler than Word Sense Disambiguation, which identifies a specific meaning for each word. The small set of concepts allows machine learning algorithms to achieve good performance when coping with the problem of tagging. However, machine learning algorithms suffer from data-sparseness. This problem becomes more evident when lexical features are involved, because test data can contain words with low frequency (or completely absent) in training data. To overcome the sparseness problem, this paper proposes a supervised method for super-sense tagging which incorporates information coming from a distributional space of words built on a large corpus. Results obtained on two standard datasets, SemCor and SensEval-3, show the effectiveness of our approach.

This paper investigates the role of Distributional Semantic Models (DSMs) in Question Answering (QA), and specifically in a QA system called Question Cube. Question Cube is a framework for QA that combines several techniques to retrieve passages containing the exact answers for natural language questions. It exploits Information Retrieval models to seek candidate answers and Natural Language Processing algorithms for the analysis of questions and candidate answers both in English and Italian. The data source for the answer is an unstructured text document collection stored in search indices. In this paper we propose to exploit DSMs in the Question Cube framework. In DSMs words are represented as mathematical points in a geometric space, also known as semantic space. Words are similar if they are close in that space. Our idea is that DSMs approaches can help to compute relatedness between users' questions and candidate answers by exploiting paradigmatic relations between words. Results of an experimental evaluation carried out on CLEF2010 QA dataset, prove the effectiveness of the proposed approach.

This paper proposes two approaches to compositional semantics in distributional semantic spaces. Both approaches conceive the semantics of complex structures, such as phrases or sentences, as being other than the sum of its terms. Syntax is the plus used as a glue to compose words. The former kind of approach encodes information about syntactic dependencies directly into distributional spaces, the latter exploits compositional operators reflecting the syntactic role of words. We present a preliminary evaluation performed on GEMS 2011 "Compositional Semantics" dataset, with the aim of understanding the effects of these approaches when applied to simple word pairs of the kind Noun-Noun, Adjective-Noun and Verb-Noun. Experimental results corroborate our conjecture that exploiting syntax can lead to improved distributional models and compositional operators, and suggest new openings for future uses in real-application scenario.

A number of works have shown that the aggregation of several Information Retrieval (IR) systems works better than each system working individually. Nevertheless, early investigation in the context of CLEF Robust-WSD task, in which semantics is involved, showed that aggregation strategies achieve only slight improvements. This paper proposes a re-ranking approach which relies on inter-document similarities. The novelty of our idea is twofold: the output of a semantic based IR system is exploited to re-weigh documents and a new strategy based on Semantic Vectors is used to compute inter-document similarities.

This paper presents the participation of the semantic Nlevels search engine SENSE at the CLEF 2009 Ad Hoc Robust-WSD Task. Our aim is to demonstrate that the combination of the N-levels model and WSD can improve the retrieval performance even when an effective retrieval model is adopted. To reach this aim, we worked on two different strategies. On one hand a model, based on Okapi BM25, was adopted at each level. On the other hand, we integrated a local relevance feedback technique, called Local Context Analysis, in both indexing levels of the system (keyword and word meaning). The hypothesis that Local Context Analysis can be effective even when it works on word meanings coming from a WSD algorithm is supported by experimental results. In monolingual task MAP increased of about 2% exploiting disambiguation, while GMAP increased from 4% to 9% when we used WSD in both mono- and bi- lingual tasks.

This paper proposes an Information Retrieval (IR) system that integrates sense discrimination to overcome the problem of word ambiguity. Word ambiguity is a key problem for systems that have access to textual information. Semantic Vectors are able to divide the usages of a word into different meanings, by discriminating among word meanings on the ground of information available in unannotated corpora. This paper has a twofold goal: the former is to evaluate the effectiveness of an IR system based on Semantic Vectors, the latter is to describe how they have been integrated in a semantic IR framework to build semantic spaces of words and documents. To achieve the first goal, we performed an in vivo evaluation in an IR scenario and we compared the method based on sense discrimination to a method based on Word Sense Disambiguation (WSD). Contrarily to sense discrimination, which aims to discriminate among different meanings not necessarily known a priori, WSD is the task of selecting a sense for a word from a set of predefined possibilities. To accomplish the second goal, we integrated Semantic Vectors in a semantic search engine called SENSE (SEmantic N-levels Search Engine).

Traditional Information Retrieval (IR) systems are based on bag-of-words representation. This approach retrieves relevant documents by lexical matching between query and document terms. Due to synonymy and polysemy, lexical methods produce imprecise or incomplete results. In this paper we present SENSE (SEmantic N-levels Search Engine), an IR system that tries to overcome the limitations of the ranked keyword approach, by introducing semantic levels which integrate (and not simply replace) the lexical level represented by keywords. Semantic levels provide information about word meanings, as described in a reference dictionary, and named entities. This paper focuses on the named entity level. Our aim is to prove that named entities are useful to improve retrieval performance. We exploit a model able to capture entity relationships, although they are not explicit in documents text. Experiments on CLEF dataset prove the effectiveness of our hypothesis.

Polysemous words have more than one possible meaning, thus word ambiguity is a key issue for the systems which access textual information. Computational linguistics proposes two main methods to cope with word ambiguity: sense disambiguation and sense discrimination. (Word) Sense Disambiguation is the task of selecting a sense for a word from a set of predefined possibilities, while (Word) Sense Discrimination is the task of dividing the usages of a word into different meanings, discriminating among word meanings based on information found in unannotated corpora. This paper proposes a strategy to compare disambiguation and discrimination systems by adopting an "in vivo" evaluation in an Information Retrieval scenario. The goal of the evaluation is to establish how disambiguation and discrimination bias the retrieval performance.

This paper presents the results of the experiments conducted at the University of Bari for the Ad Hoc Robust-WSD track of the Cross-Language Evaluation Forum (CLEF) 2008. The evaluation was performed using SENSE (SEmantic N-levels Search Engine), a semantic search engine that tries to overcome the limitations of the ranked keyword approach by introducing semantic levels, which integrate (and not simply replace) the lexical level represented by keywords. We show how SENSE is able to manage documents indexed at two separate levels, keyword and word meaning, in an attempt of improving the retrieval performance. Two types of experiments have been performed by exploiting both only one indexing level and all indexing levels at the same time. The experiments performed combining keywords and word meanings, extracted from the WordNet lexical database, show the promise of the idea and point out the value of our institution. In particular the results confirm our hypothesis: The combination of two indexing levels outperforms a single level. Indeed, an improvement of 35% in precision has been obtained by adopting the N-levels model with respect to the results obtained by exploiting the indexing level based only on keywords.

In this paper, we propose several methods for the diachronic analysis of the Italian language. We build several models by exploiting Temporal Random Indexing and the Google Ngram dataset for the Italian language. Each proposed method is evaluated on the ability to automatically identify meaning shift over time. To this end, we introduce a new dataset built by looking at the etymological information reported in some dictionaries. In questo lavoro proponiamo alcuni metodi per l’analisi diacronica della lingua italiana. Abbiamo costruito differenti modelli utilizzando la tecnica del Temporal Random Indexing e Google Ngram per l’italiano. Ciascun metodo proposto è stato valutato rispetto alla capacità di identificare automaticamente i cambi di significato nel tempo. A tale scopo introduciamo uno nuovo dataset costruito mediante le informazioni etimologiche presenti in alcuni dizionari.

This paper describes the UNIBA team participation in the Cross-Level Semantic Similarity task at SemEval 2014. We propose to combine the output of different semantic similarity measures which exploit Word Sense Disambiguation and Distributional Semantic Models, among other lexical features. The integration of similarity measures is performed by means of two supervised methods based on Gaussian Process and Support Vector Machine. Our systems obtained very encouraging results, with the best one ranked 6 th out of 38 submitted systems.

This report describes the main outcomes of the 2016 Named Entity rEcognition and Linking in Italian Tweet (NEEL-IT) Challenge. The goal of the challenge is to provide a benchmark corpus for the evaluation of entity recognition and linking algorithms specifically designed for noisy and short texts, like tweets, written in Italian. The task requires the correct identification of entity mentions in a text and their linking to the proper named entities in a knowledge base. To this aim, we choose to use the canonicalized dataset of DBpedia 2015-10. The task has attracted five participants, for a total of 15 runs submitted. In questo report descriviamo i principali risultati conseguiti nel primo task per la lingua Italiana di Named Entity rEcognition e Linking in Tweet (NEEL-IT). Il task si prefigge l’obiettivo di offrire un framework di valutazione per gli algoritmi di riconoscimento e linking di entità a nome proprio specificamente disegnati per la lingua italiana per testi corti e rumorosi, quali i tweet. Il task si compone di una fase di riconoscimento delle menzioni di entità con nome proprio nel testo e del loro successivo collegamento alle opportune entità in una base di conoscenza. In questo task abbiamo scelto come base di conoscenza la versione canonica di DBpedia 2015. Il task ha attirato cinque partecipanti per un totale di 15 diversi run.