<link>https://example.com/publications/</link></image><item><title>Adapting Language Models with Continual Learning for Temporal Drifts</h1> <article> <h1>Adapting Language Models with Continual Learning for Temporal Drifts</h1> <p>Thu, 01 Jan 2026 00:00:00 +0000</p> </article> <article> <h1>An Experimental Comparison of the Most Popular Approaches to Fake News Detection</h1> <p>Thu, 01 Jan 2026 00:00:00 +0000</p> </article> <article> <h1>Challenging the Abilities of Large Language Models in Italian: A Community Initiative</h1> <p>Thu, 01 Jan 2026 00:00:00 +0000</p> </article> <article> <h1>Design Pattern-Based Code Refactoring with Llms</h1> <p>Thu, 01 Jan 2026 00:00:00 +0000</p> </article> <article> <h1>Emotion Recognition in Multimodal Social Data</h1> <p>Thu, 01 Jan 2026 00:00:00 +0000</p> <h2 id="abstract">Abstract</h2> <p>Emotion recognition on social media is often approached in unimodal or single-label settings, despite the multimodal nature of online communication. This paper presents a study of multilabel emotion recognition from paired text-image data. We evaluate vision–language encoders and compare them with strong unimodal baselines and a zero-shot multimodal LLM. A simple multimodal classifier built on CLIP achieves the most reliable performance. Data-centric additions such as emoji transcription, caption augmentation, and pseudo-labelling offer limited gains, whereas calibrated decision thresholds have a consistent effect. The results highlight the value of visual cues and show limitations of recent VLMs.</p> </article> <article> <h1>Enhancing Debunking Effectiveness through LLM-based Personality Adaptation</h1> <p>Thu, 01 Jan 2026 00:00:00 +0000</p> </article> <article> <h1>Evaluating Online Moderation via LLM-powered Counterfactual Simulations</h1> <p>Thu, 01 Jan 2026 00:00:00 +0000</p> <h2 id="abstract">Abstract</h2> <p>Online Social Networks (OSNs) widely adopt content moderation to mitigate the spread of abusive and toxic discourse. Nonetheless, the real effectiveness of moderation interventions remains unclear due to the high cost of data collection and limited experimental control. The latest developments in Natural Language Processing pave the way for a new evaluation approach. Large Language Models (LLMs) can be successfully leveraged to enhance Agent-Based Modeling and simulate human-like social behavior with unprecedented degree of believability. Yet, existing tools do not support simulation-based evaluation of moderation strategies. We fill this gap by designing a LLM-powered simulator of OSN conversations enabling a parallel, counterfactual simulation where toxic behavior is influenced by moderation interventions, keeping all else equal. We conduct extensive experiments, unveiling the psychological realism of OSN agents, the emergence of social contagion phenomena and the superior effectiveness of personalized moderation strategies.</p> </article> <article> <h1>RAG-enhanced Llms for Interactive Explainability in Clinical Decision Support Systems</h1> <p>Thu, 01 Jan 2026 00:00:00 +0000</p> </article> <article> <h1>Standard vs. Modular Sampling: Best Practices for Reliable LLM Unlearning</h1> <p>Thu, 01 Jan 2026 00:00:00 +0000</p> </article> <article> <h1>All-in-One: Understanding and Generation in Multimodal Reasoning with the MAIA Benchmark</h1> <p>Wed, 01 Jan 2025 00:00:00 +0000</p> <h2 id="abstract">Abstract</h2> <p>We introduce MAIA (Multimodal AI Assessment), a native-Italian benchmark designed for fine-grained investigation of the reasoning abilities of visual language models on videos. MAIA differs from other available video benchmarks for its design, its reasoning categories, the metric it uses, and the language and culture of the videos. MAIA evaluates Vision Language Models (VLMs) on two aligned tasks: a visual statement verification task, and an open-ended visual question-answering task, both on the same set of video-related questions. It considers twelve reasoning categories that aim to disentangle language and vision relations by highlighting the role of the visual input. Thanks to its carefully taught design, it evaluates VLMs’ consistency and visually grounded natural language comprehension and generation simultaneously through an aggregated metric revealing low results that highlight models’ fragility. Last but not least, the video collection has been carefully selected to reflect the Italian culture, and the language data are produced by native-speakers.</p> </article> <article> <h1>Call2Go: A Cross-Platform Strategy of Kids Online Social Media Influencers</h1> <p>Wed, 01 Jan 2025 00:00:00 +0000</p> </article> <article> <h1>Continually Learn to Map Visual Concepts to Language Models in Resource-Constrained Environments</h1> <p>Wed, 01 Jan 2025 00:00:00 +0000</p> </article> <article> <h1>Embracing Diversity: A Multi-Perspective Approach with Soft Labels</h1> <p>Wed, 01 Jan 2025 00:00:00 +0000</p> <h2 id="abstract">Abstract</h2> <p>In subjective tasks like stance detection, diverse human perspectives are often simplified into a single ground truth through label aggregation i.e. majority voting, potentially marginalizing minority viewpoints. This paper presents a Multi-Perspective framework for stance detection that explicitly incorporates annotation diversity by using soft labels derived from both human and large language model (LLM) annotations. Building on a stance detection dataset focused on controversial topics, we augment it with document summaries and new LLM-generated labels. We then compare two approaches: a baseline using aggregated hard labels, and a multi-perspective model trained on disaggregated soft labels that capture annotation distributions. Our findings show that multi-perspective models consistently outperform traditional baselines (higher F1-scores), with lower model confidence, reflecting task subjectivity. This work highlights the importance of modeling disagreement and promotes a shift toward more inclusive, perspective-aware NLP systems.</p> </article> <article> <h1>ExpliCa: Evaluating Explicit Causal Reasoning in Large Language Models</h1> <p>Wed, 01 Jan 2025 00:00:00 +0000</p> </article> <article> <h1>Investigating Time-Scales in Deep Echo State Networks for Natural Language Processing</h1> <p>Wed, 01 Jan 2025 00:00:00 +0000</p> </article> <article> <h1>Perspectives in Play: A Multi-Perspective Approach for More Inclusive NLP Systems</h1> <p>Wed, 01 Jan 2025 00:00:00 +0000</p> </article> <article> <h1>Prompting Encoder Models for Zero-Shot Classification: A Cross-Domain Study in Italian</h1> <p>Wed, 01 Jan 2025 00:00:00 +0000</p> </article> <article> <h1>Challenging Specialized Transformers on Zero-Shot Classification</h1> <p>Mon, 01 Jan 2024 00:00:00 +0000</p> <h2 id="abstract">Abstract</h2> <p>This paper investigates the feasibility of employing basic prompting systems for domain-specific language models. The study focuses on bureaucratic language and uses the recently introduced BureauBERTo model for experimentation. The experiments reveal that while further pre-trained models exhibit reduced robustness concerning general knowledge, they display greater adaptability in modeling domain-specific tasks, even under a zero-shot paradigm. This demonstrates the potential of leveraging simple prompting systems in specialized contexts, providing valuable insights both for research and industry.</p> </article> <article> <h1>Characterizing Fashion Influencers' Behavior on Instagram</h1> <p>Mon, 01 Jan 2024 00:00:00 +0000</p> </article> <article> <h1>Continual Pre-Training Mitigates Forgetting in Language and Vision</h1> <p>Mon, 01 Jan 2024 00:00:00 +0000</p> <h2 id="abstract">Abstract</h2> <p>: Pre-trained models are commonly used in Continual Learning to initialize the model before training on the stream of non-stationary data. However, pre-training is rarely applied during Continual Learning. We investigate the characteristics of the Continual Pre-Training scenario, where a model is continually pre-trained on a stream of incoming data and only later fine-tuned to different downstream tasks. We introduce an evaluation protocol for Continual Pre-Training which monitors forgetting against a Forgetting Control dataset not present in the continual stream. We disentangle the impact on forgetting of 3 main factors: the input modality (NLP, Vision), the architecture type (Transformer, ResNet) and the pre-training protocol (supervised, self-supervised). Moreover, we propose a Sample-Efficient Pre-training method (SEP) that speeds up the pre-training phase. We show that the pre-training protocol is the most important factor accounting for forgetting. Surprisingly, we discovered that self-supervised continual pre-training in both NLP and Vision is sufficient to mitigate forgetting without the use of any Continual Learning strategy. Other factors, like model depth, input modality and architecture type are not as crucial.</p> </article> <article> <h1>Dataset for Multimodal Fake News Detection and Verification Tasks</h1> <p>Mon, 01 Jan 2024 00:00:00 +0000</p> </article> <article> <h1>Deep Continual Learning for Medical Call Incidents Text Classification under the Presence of Dataset Shifts</h1> <p>Mon, 01 Jan 2024 00:00:00 +0000</p> <h2 id="abstract">Abstract</h2> <p>The aim of this work is to develop and evaluate a deep classifier that can effectively prioritize Emergency Medical Call Incidents (EMCI) according to their life-threatening level under the presence of dataset shifts. We utilized a dataset consisting of 1982746 independent EMCI instances obtained from the Health Services Department of the Region of Valencia (Spain), with a time span from 2009 to 2019 (excluding 2013). The dataset includes free text dispatcher observations recorded during the call, as well as a binary variable indicating whether the event was life-threatening. To evaluate the presence of dataset shifts, we examined prior probability shifts, covariate shifts, and concept shifts. Subsequently, we designed and implemented four deep Continual Learning (CL) strategies-cumulative learning, continual fine-tuning, experience replay, and synaptic intelligence-alongside three deep CL baselines-joint training, static approach, and single fine-tuning-based on DistilBERT models. Our results demonstrated evidence of prior probability shifts, covariate shifts, and concept shifts in the data. Applying CL techniques had a statistically significant ($\alpha$=0.05) positive impact on both backward and forward knowledge transfer, as measured by the F1-score, compared to non-continual approaches. We can argue that the utilization of CL techniques in the context of EMCI is effective in adapting deep learning classifiers to changes in data distributions, thereby maintaining the stability of model performance over time. To our knowledge, this study represents the first exploration of a CL approach using real EMCI data.</p> </article> <article> <h1>Investigating the Hurtfulness of Misogynistic Tweets across Professions</h1> <p>Mon, 01 Jan 2024 00:00:00 +0000</p> </article> <article> <h1>Multi-Perspective Stance Detection</h1> <p>Mon, 01 Jan 2024 00:00:00 +0000</p> </article> <article> <h1>RoBEXedda: Sexism Detection in Tweets</h1> <p>Mon, 01 Jan 2024 00:00:00 +0000</p> </article> <article> <h1>Updating Knowledge in Large Language Models: An Empirical Evaluation</h1> <p>Mon, 01 Jan 2024 00:00:00 +0000</p> </article> <article> <h1>BureauBERTo: Adapting UmBERTo to the Italian Bureaucratic Language</h1> <p>Sun, 01 Jan 2023 00:00:00 +0000</p> <h2 id="abstract">Abstract</h2> <p>In this work, we introduce BureauBERTo, the first transformer-based language model adapted to the Italian Public Administration (PA) and technical-bureaucratic domains. We further pre-trained the general-purpose Italian model UmBERTo on a corpus of PA, banking, and insurance documents, and we expanded UmBERTo’s vocabulary with domain-specific terms. We show that BureauBERTo benefitted from the adaptation by comparing it with UmBERTo in both an intrinsic and extrinsic evaluation. The intrinsic evaluation has been conducted through specific fill-mask experiments. The extrinsic one has been faced with a named entity recognition task on one of the sub-domains in BureauBERTo.</p> </article> <article> <h1>Challenging Specialized Transformers on Zero-Shot Classification</h1> <p>Sun, 01 Jan 2023 00:00:00 +0000</p> <h2 id="abstract">Abstract</h2> <p>This paper investigates the feasibility of employing basic prompting systems for domain-specific language models. The study focuses on bureaucratic language and uses the recently introduced BureauBERTo model for experimentation. The experiments reveal that while further pre-trained models exhibit reduced robustness concerning general knowledge, they display greater adaptability in modeling domain-specific tasks, even under a zero-shot paradigm. This demonstrates the potential of leveraging simple prompting systems in specialized contexts, providing valuable insights both for research and industry.</p> </article> <article> <h1>CheckIT!: A Corpus of Expert Fact-Checked Claims for Italian</h1> <p>Sun, 01 Jan 2023 00:00:00 +0000</p> <h2 id="abstract">Abstract</h2> <p>This paper introduces CheckIT!, a resource of expert fact-checked claims, filling a gap for the development of fact-checking pipelines in Italian. We further investigate the use of three state-of-the-art generative text models to create variations of claims in zero-shot settings as a data-augmentation strategy for the identification of previously fact-checked claims. Our results indicate that models struggles in varying the surface forms of the claims.</p> </article> <article> <h1>Inconsistency Detection in Natural Language Requirements Using ChatGPT: A Preliminary Evaluation</h1> <p>Sun, 01 Jan 2023 00:00:00 +0000</p> </article> <article> <h1>MULTI-fake-DetectiVE at EVALITA 2023: Overview of the Multimodal Fake News Detection and Verification Task</h1> <p>Sun, 01 Jan 2023 00:00:00 +0000</p> </article> <article> <h1>MULTI-fake-DetectiVE at EVALITA 2023: Overview of the Multimodal Fake News Detection and Verification Task</h1> <p>Sun, 01 Jan 2023 00:00:00 +0000</p> <h2 id="abstract">Abstract</h2> <p>This paper introduces the MULTI-Fake-DetectiVE shared task for the EVALITA 2023 campaign. The task was aimed at exploring multimodality within the realm of fake news and intended to address the problem from two perspectives, represented by the two sub-tasks. In sub-task 1, we aimed to evaluate the effectiveness of multimodal fake news detection systems. In sub-task 2, we sought to gain insights into the interplay between text and images, specifically how they mutually influence the interpretation of content in the context of distinguishing between fake and real news. Both perspectives were framed as classification problems. The paper presents an overview of the task. In particular, we detail the key aspects of the task, including the creation of a new dataset for fake news detection in Italian, the evaluation methodology and criteria, the participant systems, and their results. In light of the obtained results, we argue that the problem is still open and propose some future directions.</p> </article> <article> <h1>Special Issue NL4AI 2022: Workshop on Natural Language for Artificial Intelligence</h1> <p>Sun, 01 Jan 2023 00:00:00 +0000</p> </article> <article> <h1>WiC-ITA at EVALITA2023: Overview of the EVALITA2023 Word-in-Context for Italian Task</h1> <p>Sun, 01 Jan 2023 00:00:00 +0000</p> </article> <article> <h1>WiC-ITA at EVALITA2023: Overview of the EVALITA2023 Word-in-Context for Italian Task</h1> <p>Sun, 01 Jan 2023 00:00:00 +0000</p> </article> <article> <h1>Bias Discovery within Human Raters: A Case Study of the Jigsaw Dataset</h1> <p>Sat, 01 Jan 2022 00:00:00 +0000</p> </article> <article> <h1>Evaluating Pre-Trained Transformers on Italian Administrative Texts</h1> <p>Sat, 01 Jan 2022 00:00:00 +0000</p> </article> <article> <h1>GQA-it: Italian Question Answering on Image Scene Graphs</h1> <p>Sat, 01 Jan 2022 00:00:00 +0000</p> </article> <article> <h1>GQA-it: Italian Question Answering on Image Scene Graphs</h1> <p>Sat, 01 Jan 2022 00:00:00 +0000</p> </article> <article> <h1>In-Context Annotation of Topic-Oriented Datasets of Fake News: A Case Study on the Notre-Dame Fire Event</h1> <p>Sat, 01 Jan 2022 00:00:00 +0000</p> <h2 id="abstract">Abstract</h2> <p>The problem of fake news detection is becoming increasingly interesting for several research fields. Different approaches have been proposed, based on either the content of the news itself or the context and properties of its spread over time, specifically on social media. In the literature, it does not exist a widely accepted general-purpose dataset for fake news detection, due to the complexity of the task and the increasing ability to produce fake news appearing credible in particular moments. In this paper, we propose a methodology to collect and label news pertinent to specific topics and subjects. Our methodology focuses on collecting data from social media about real-world events which are known to trigger fake news. We propose a labelling method based on crowdsourcing that is fast, reliable, and able to approximate expert human annotation. The proposed method exploits both the content of the data (i.e., the texts) and contextual information about fake news for a particular real-world event. The methodology is applied to collect and annotate the Notre-Dame Fire Dataset and to annotate part of the PHEME dataset. Evaluation is performed with fake news classifiers based on Transformers and fine-tuning. Results show that context-based annotation outperforms traditional crowdsourcing out-of-context annotation.</p> </article> <article> <h1>MATE, a Meta Layer between Natural Language and Database</h1> <p>Sat, 01 Jan 2022 00:00:00 +0000</p> </article> <article> <h1>Preface to the Sixth Workshop on Natural Language for Artificial Intelligence (NL4AI)</h1> <p>Sat, 01 Jan 2022 00:00:00 +0000</p> </article> <article> <h1>Leveraging CLIP for Image Emotion Recognition</h1> <p>Fri, 01 Jan 2021 00:00:00 +0000</p> <h2 id="abstract">Abstract</h2> <p>Multi-modal neural models that are able to encode and process both visual and textual data are becoming more and more common in the last few years. Such models enable new ways to learn the interaction between vision and text, and thus can be successfully applied to tasks of varying complexity in the domain of image and text classification. However, such models are traditionally oriented to learn grounded properties of images and of the objects they depict and less suited to solve tasks involving subjective characteristics, such as the emotions they can convey in viewers. In this paper, we provide some insights in the performances of the recently released OpenAI CLIP model for an emotion classification task. We evaluate the model both under zero-shot settings and via fine tuning on an image-emotion dataset. We compare the performances of CLIP both in a zero-shot and fine-tuning setting on (i) a standard benchmark dataset for object recognition (ii) an image-emotion dataset. Moreover, we evaluate to which extent a CLIP model adapted to emotions is able to retain general knowledge and generalization capabilities.</p> </article> <article> <h1>Preface to the Fifth Workshop on Natural Language for Artificial Intelligence (NL4AI)</h1> <p>Fri, 01 Jan 2021 00:00:00 +0000</p> <h2 id="abstract">Abstract</h2> <p>Preface to the fifth Workshop on Natural Language for Artificial Intelligence (NL4AI)</p> </article> <article> <h1>CAPISCO@CONcreTEXT 2020: (Un)Supervised Systems to Contextualize Concreteness with Norming Data</h1> <p>Wed, 01 Jan 2020 00:00:00 +0000</p> </article> <article> <h1>EVALITA 2020: Overview of the 7th Evaluation Campaign of Natural Language Processing and Speech Tools for Italian</h1> <p>Wed, 01 Jan 2020 00:00:00 +0000</p> </article> <article> <h1>EVALITA 2020: Overview of the 7th Evaluation Campaign of Natural Language Processing and Speech Tools for Italian</h1> <p>Wed, 01 Jan 2020 00:00:00 +0000</p> <h2 id="abstract">Abstract</h2> <p>The Evaluation Campaign of Natural Language Processing and Speech Tools for Italian (EVALITA) is the biennial initiative aimed at promoting the development of language and speech technologies for the Italian language. EVALITA is promoted by the Italian Association of Computational Linguistics (AILC) and it is endorsed by the Italian Association for Artificial Intelligence (AIxIA) and the Italian Association for Speech Sciences (AISV). EVALITA provides a shared framework where different systems and approaches can be scientifically evaluated and compared with each other with respect to a large variety of tasks, suggested and organized by the Italian research community. The proposed tasks represent scientific challenges where methods, resources, and systems can be tested against shared benchmarks representing linguistic open issues or real world applications, possibly in a multilingual and/or multi-modal perspective. The collected data sets provide big opportunities for scientists to explore old and new problems concerning NLP in Italian as well as to develop solutions and to discuss the NLP-related issues within the community. Some tasks are traditionally present in the evaluation campaign, while others are completely new. This paper introduces the tasks proposed at EVALITA 2020 and provides an overview to the participants and systems whose descriptions and obtained results are reported in these Proceedings.</p> </article> <article> <h1>FRAQUE: A Frame-Based Question-Answering System for the Public Administration Domain</h1> <p>Wed, 01 Jan 2020 00:00:00 +0000</p> </article> <article> <h1>Less Is MORE: A MultimOdal System for Tag Refinement</h1> <p>Wed, 01 Jan 2020 00:00:00 +0000</p> </article> <article> <h1>Lessons Learned from EVALITA 2020 and Thirteen Years of Evaluation of Italian Language Technology</h1> <p>Wed, 01 Jan 2020 00:00:00 +0000</p> </article> <article> <h1>Preface to the EVALITA 2020 Proceedings</h1> <p>Wed, 01 Jan 2020 00:00:00 +0000</p> </article> <article> <h1>UNIPI-NLE at CheckThat! 2020: Approaching Fact Checking from a Sentence Similarity Perspective through the Lens of Transformers</h1> <p>Wed, 01 Jan 2020 00:00:00 +0000</p> <h2 id="abstract">Abstract</h2> <p>This paper describes a Fact Checking system based on a combination of Information Extraction and Deep Learning strategies to approach the task named Verified Claim Retrieval" (Task 2) for the CheckThat! 2020 evaluation campaign. The system is based on two main assumptions: a claim that verifies a tweet is expected i) to mention the same entities and keyphrases, and ii) to have a similar meaning. The former assumption has been addressed by exploiting an Information Extraction module capable of determining the pairs in which the tweet and the claim share at least a named entity or a relevant keyword. To address the latter, we exploited Deep Learning to refine the computation of the text similarity between a tweet and a claim, and to actually classify the pairs as correct matches or not. In particular, the system has been built starting from a pre-trained Sentence-BERT model, on which two cascade fine-tuning steps have been applied in order to i) assign a higher cosine similarity to gold pairs, and ii) classify a pair as correct or not. The final ranking produced by the system is the probability of the pair labelled as correct. Overall, the system reached a 0.91 MAP@5 on the test set.</p> </article> <article> <h1>Voices of the Great War: A Richly Annotated Corpus of Italian Texts on the First World War</h1> <p>Wed, 01 Jan 2020 00:00:00 +0000</p> </article> <article> <h1>Do Idioms Have a Heart? The SIDE (Sentiment of Idiomatic Expressions) Project</h1> <p>Tue, 01 Jan 2019 00:00:00 +0000</p> </article> <article> <h1>Text Frame Detector: Slot Filling Based on Domain Knowledge Bases</h1> <p>Tue, 01 Jan 2019 00:00:00 +0000</p> </article> <article> <h1>CoreNLP-it: A UD Pipeline for Italian Based on Stanford CoreNLP</h1> <p>Mon, 01 Jan 2018 00:00:00 +0000</p> </article> <article> <h1>SemplicePA Semantic Instruments for PubLIc Administrators and CitizEns</h1> <p>Mon, 01 Jan 2018 00:00:00 +0000</p> </article> <article> <h1>The Emotions of Abstract Words: A Distributional Semantic Analysis</h1> <p>Mon, 01 Jan 2018 00:00:00 +0000</p> </article> <article> <h1>Voci Della Grande Guerra Preserving the Digital Memory of World War I</h1> <p>Mon, 01 Jan 2018 00:00:00 +0000</p> </article> <article> <h1>Emo2Val: Inferring Valence Scores from Fine-Grained Emotion Values</h1> <p>Sun, 01 Jan 2017 00:00:00 +0000</p> </article> <article> <h1>How to Harvest Word Combinations from Corpora. Methods, Evaluation and Perspectives</h1> <p>Sun, 01 Jan 2017 00:00:00 +0000</p> </article> <article> <h1>INFORMed PA: A NER for the Italian Public Administration Domain</h1> <p>Sun, 01 Jan 2017 00:00:00 +0000</p> </article> <article> <h1>Learning Affect with Distributional Semantic Models</h1> <p>Sun, 01 Jan 2017 00:00:00 +0000</p> </article> <article> <h1>Evaluating Context Selection Strategies to Build Emotive Vector Space Models</h1> <p>Fri, 01 Jan 2016 00:00:00 +0000</p> </article> <article> <h1>Exploiting Emotive Features for the Sentiment Polarity Classification of Tweets</h1> <p>Fri, 01 Jan 2016 00:00:00 +0000</p> </article> <article> <h1>Extracting Terms with Extra</h1> <p>Fri, 01 Jan 2016 00:00:00 +0000</p> </article> <article> <h1>FB-NEWS15: A Topic-Annotated Facebook Corpus for Emotion Detection and Sentiment Analysis</h1> <p>Fri, 01 Jan 2016 00:00:00 +0000</p> </article> <article> <h1>Pos-Patterns or Syntax? Comparing Methods for Extracting Word Combinations</h1> <p>Fri, 01 Jan 2016 00:00:00 +0000</p> </article> <article> <h1>Automatic Extraction of Word Combinations from Corpora: Evaluating Methods and Benchmarks</h1> <p>Thu, 01 Jan 2015 00:00:00 +0000</p> <h2 id="abstract">Abstract</h2> <p>We report on three experiments aimed at comparing two popular methods for the automatic extraction of Word Combinations from corpora, with a view to evaluate: i) their efficacy in acquiring data to be included in a combinatory resource for Italian; ii) the impact of different types of benchmarks on the evaluation itself.</p> </article> <article> <h1>ItEM: A Vector Space Model to Bootstrap an Italian Emotive Lexicon</h1> <p>Thu, 01 Jan 2015 00:00:00 +0000</p> <h2 id="abstract">Abstract</h2> <p>In recent years computational linguistics has seen a rising interest in subjectivity, opinions, feelings and emotions. Even though great attention has been given to polarity recognition, the research in emotion detection has had to rely on small emotion resources. In this paper, we present a methodology to build emotive lexicons by jointly exploiting vector space models and human annotation, and we provide the first results of the evaluation with a crowdsourcing experiment.</p> </article> <article> <h1>``Il Piave Mormorava...'': Recognizing Locations and Other Named Entities in Italian Texts on the Great War</h1> <p>Wed, 01 Jan 2014 00:00:00 +0000</p> </article> <article> <h1>Computational Analysis of Historical Documents: An Application to Italian War Bulletins in World War I and II</h1> <p>Wed, 01 Jan 2014 00:00:00 +0000</p> <h2 id="abstract">Abstract</h2> <p>World War (WW) I and II represent crucial landmarks in the history on mankind: They have affected the destiny of whole generations and their consequences are still alive throughout Europe. In this paper we present an ongoing project to carry out a computational analysis of Italian war bulletins in WWI and WWII, by applying state-of-the-art tools for NLP and Information Extraction. The annotated texts and extracted information will be explored with a dedicated Web interface, allowing for multidimensional access and exploration of historical events through space and time.</p> </article> <article> <h1>The CoLing Lab System for Sentiment Polarity Classification of Tweets</h1> <p>Wed, 01 Jan 2014 00:00:00 +0000</p> <h2 id="abstract">Abstract</h2> <p>This paper describes the CoLing Lab system for the EVALITA 2014 SENTIment POLarity Classification (SENTIPOLC) task. Our system is based on a SVM classifier trained on the rich set of lexical, global and twitter-specific features described in these pages. Overall, our system reached a 0.63 weighted F-score on the test set provided by the task organizers.</p> </article> </main></body></html>