Despite all the fact-checking initiatives on news and social media aimed at countering misinformation, they remain insufficient to promptly address the wide array of misleading information disseminated by both news and social media outlets. Rather than attempting to identify or filter misleading information, this work advocates new tools for assisting online readers in identifying misinformation among the massive online content pushed every day through multiple platforms. We introduce DOMAIN, an article assessment resource bundle comprising a multidimensional indicator to categorize articles into different types (hard news, soft news, opinion, satire, and conspiracy), a set of explanatory metrics to help users understand the results, a tool for verifying the reliability of the article’s source, and a text summary of the assessment. This work also studies how DOMAIN tools impact online readers, specifically focusing on (i) understanding the extent to which computer-generated assessments influence human perceptions of credibility; (ii) evaluating the effectiveness of automatic article categorization in human assessment of credibility; and (iii) identifying the most relevant explanatory metrics for promoting informed and critical consumption of information.

Our society is a complex network of people that requires cooperation and socialization to thrive, being that interaction and communication between individuals have been key parts of our society since the beginning of its existence. With the rise of the Covid-19 pandemic and the need to quarantine, this complex need has been truly affected on many sectors, and we became less connected. Technology was an important factor during the early days of the pandemic, with people starting to opt for certain systems and platforms that allowed them to stay in touch with each other, however nowadays, there are still massive limitations to these platforms when it comes to an interaction’s ambiance due to the lack of user immersion. In this paper, we present an MMORPG with a scalable infrastructure able to host events for hundreds of people, but most importantly that allows users to communicate in a more realistic manner with a set of functionalities that fulfil the goal to combat the undesirable isolation. To verify the effectiveness of our platform, questionnaires, and usability tests were conducted.

We introduce a deep neural network for automated sarcasm detection.Recent work has emphasized the need for models to capitalize on contextual features, beyond lexical and syntactic cues present in utterances.For example, different speakers will tend to employ sarcasm regarding different subjects and, thus, sarcasm detection models ought to encode such speaker information.Current methods have achieved this by way of laborious feature engineering.By contrast, we propose to automatically learn and then exploit user embeddings, to be used in concert with lexical signals to recognize sarcasm.Our approach does not require elaborate feature engineering (and concomitant data scraping); fitting user embeddings requires only the text from their previous posts.The experimental results show that the our model outperforms a state-of-the-art approach leveraging an extensive set of carefully crafted features.

Ramon F. Astudillo, Silvio Amir, Wang Ling, Mário Silva, Isabel Trancoso. Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). 2015.

Semi-supervised bootstrapping techniques for relationship extraction from text iteratively expand a set of initial seed relationships while limiting the semantic drift.We research bootstrapping for relationship extraction using word embeddings to find similar relationships.Experimental results show that relying on word embeddings achieves a better performance on the task of extracting four types of relationships from a collection of newswire documents when compared with a baseline using TF-IDF to find similar relationships.

Web archives already hold together more than 534 billion files and this number continues to grow as new initiatives arise. Searching on all versions of these files acquired throughout time is challenging, since users expect as fast and precise answers from web archives as the ones provided by current web search engines. This work studies, for the first time, how to improve the search effectiveness of web archives, including the creation of novel temporal features that explore the correlation found between web document persistence and relevance. The persistence was analyzed over 14 years of web snapshots. Additionally, we propose a temporal-dependent ranking framework that exploits the variance of web characteristics over time influencing ranking models. Based on the assumption that closer periods are more likely to hold similar web characteristics, our framework learns multiple models simultaneously, each tuned for a specific period. Experimental results show significant improvements over the search effectiveness of single-models that learn from all data independently of its time. Thus, our approach represents an important step forward on the state-of-the-art IR technology usually employed in web archives.

Most geographic information retrieval systems depend on the detection and disambiguation of place names in documents, assuming that the documents with a specific geographic scope contain explicit place names in the text that are strongly related to the document scopes. However, some non-geographic names such as companies, monuments or sport events, may also provide indirect relevant evidence that can significantly contribute to the assignment of geographic scopes to documents. In this paper, we analyze the amount of implicit and explicit geographic evidence in newspaper documents, and measure its impact on geographic information retrieval by evaluating the performance of a retrieval system using the GeoCLEF evaluation data.

Abstract This paper documents hazardous situations on the Web that crawlers must address. This knowledge was accumulated while developing and operating the Viúva Negra (VN) crawler to feed a search engine and a Web archive for the Portuguese Web for four years. The design, implementation and evaluation of the VN crawler are also presented as a case study of a Web crawler design. The case study tested provides crawling techniques that may be useful for the further development of crawlers. Copyright © 2007 John Wiley & Sons, Ltd.

Erroneous data can often be found in databases, and detecting it is normally a non-trivial task. For example, To cope with the large amount of biological sequences being produced, a significant number of genes and proteins have been annotated by automated tools. A protein annotation is an association between a protein and a term describing its role. These tools have produced a significant number of misannotations that are now present in biological databases. This paper proposes a new method for automatically scoring associations by comparing them to preexisting curated associations. An association is a pair that links two entities. The score can be used to filter incorrect or uncommon associations.We evaluated the method using the automated protein annotations submitted to BioCreAtIvE, an international evaluation of state-of-the-art text-mining systems in Biology. The method scored each of these annotations and those scored below a certain threshold were discarded. The results have shown a small trade-off in recall for a large improvement in precision. For example, we were able to discard 44.6%, 66.8% and 81% of the misannotations, maintaining 96.9%, 84.2%, and 47.8% of the correct annotations, respectively. Moreover, we were able to outperform each individual submission to BioCreAtIvE by proper adjustment of the threshold.

Models of web data persistency are essential tools for the designof efficient information extraction systems that repeatedlycollect and process the data. This study models the persistence ofweb data through the measurement of URL and content persistenceacross several snapshots of a national community web, collectedfor 3 years. We found that the lifetimes of URLs and contents aremodelled by logarithmic functions. We gathered statistics on thestructure of the web, identified reasons for URL death andcharacterized persistent URLs and contents. The lasting contentstend to be referenced by different URLs during their lifetime,while half of the contents referenced by persistent URLs do notchange.

Many bioinformatics applications would benefit from comparing proteins based on their biological role rather than their sequence. In most biological databases, proteins are already annotated with ontology terms. Previous studies identified a correlation between the sequence similarity and the semantic similarity of proteins. The semantic similarity of proteins was computed from their annotated GO terms. However, proteins sharing a biological role do not necessarily have a similar sequence.This paper introduces our study of the correlation between GO and family similarity. Family similarity overcomes some of the limitations of sequence similarity, thus we obtained a strong correlation between GO and family similarity. Additionally, this paper introduces GraSM, a novel method that uses all the information in the graph structure of the GO, instead of considering it as a hierarchical tree. When calculating the semantic similarity of two concepts, GraSM selects the disjunctive common ancestors rather than only using the most informative common ancestor. GraSM produced a higher family similarity correlation than the original semantic similarity measures.

This paper presents an application of PageRank for assigning documents with a corresponding geographical scope. We describe the technique in detail, together with its theoretical formulation. Experimental results are promising, comparing favorably with previous proposals.

Automated methods for resource annotation are a clear necessity, as the success of the semantic Web depends on the availability of Web resources with meta data conforming to known standards and ontologies. This paper describes the WebCAT framework for automatically generating RDF descriptions of Web pages. We present a general view of the system and the algorithms involved, giving an emphasis to typical issues in processing Web data.

BACKGROUND: The development of text mining systems that annotate biological entities with their properties using scientific literature is an important recent research topic. These systems need first to recognize the biological entities and properties in the text, and then decide which pairs represent valid annotations. METHODS: This document introduces a novel unsupervised method for recognizing biological properties in unstructured text, involving the evidence content of their names. RESULTS: This document shows the results obtained by the application of our method to BioCreative tasks 2.1 and 2.2, where it identified Gene Ontology annotations and their evidence in a set of articles. CONCLUSION: From the performance obtained in BioCreative, we concluded that an automatic annotation system can effectively use our method to identify biological properties in unstructured text.

This article presents a characterization of the community Web of the people of Portugal. We defined criteria for delimiting this Web based on our past experience of crawling pages related to Portugal and collected over 3.2 million documents from 46,000 sites satisfying those criteria. Our characterization was derived from this crawl. We describe the rules that we established for defining the boundaries of this community Web and the methodology used to gather statistics. Statistics cover the number and domain distribution of sites; the number, type and size distribution of text documents; and the linkage structure of this Web. We also show how crawling constraints and abnormal situations on the Web can influence the statistics.

We present our framework for classification of Web news, based on support vector machines, and some of the initial measurements of its accuracy.

Abstract — Most information and services required today by designers will soon become available as documents distributed in a wide area hypermedia network. New integration services are required from the design environment, supporting business transactions with design information providers, automatic exchange of design data between independent groups, and inte-grated support for new forms of collaboration. We discuss design using electronic commerce and other services based on the Inter-net, and propose a hypermedia system organization for a new generation of CAD systems, conceived to make efficient use of that infrastructure. We also describe our experience as designers of an integrated design and documentation system that interfaces existing design and documentation tools with electronic com-merce services based on the World Wide Web. I.

Abstruc&Wepresent a new system for documentation in VLSI design environments.The system integrates hypermedia and some services of a CAD applications framework to offer a radically new user interface to designers.This combination of technologies enables the use of the documentation system as a frontend to the design tools, resulting in a single interface metaphor for manipulating design data, processes and the design environment.Through the use of design data and history management services, the design pmeess becomes like writing a notebook, containing all the information associated with a design as it is being conceived.Hypermedia ptwvides new capabilities for integrating media into design documentation and for non-linear views of documents.Invoking the design tools from within the documentation system adds the capability of direct manipulation of the design objects and their re-use as part of the documentation.This paper describes the conceptual model for a design system front-end embodying these ideas and a prototype implemetttation.

This paper describes the extensions made for adding support for group development within the Oct/VEM CAD framework. In our implementation, a set of mechanisms has been incorporated into the Oct library. These contain support for versioning and concurrent access to Oct design objects. The mechanisms can be configured for establishment of specific design management styles. As an example, there is now support for organization of designs in terms of workspaces; but, the number of workspaces or the relationship between them must be established externally by a design management tool. Design management tools configure these mechanisms to establish policies to be followed by the other tools. With this architecture, integration of existing tools with distinct design management styles then becomes feasible.