arXiv subjects

Seyed Hossein Jafari

Publications and source records attributed to Seyed Hossein Jafari.

3 recordsLinked to original sources

arXiv2021

COfEE: A Comprehensive Ontology for Event Extraction from text

Data is published on the web over time in great volumes, but majority of the data is unstructured, making it hard to understand and difficult to interpret. Information Extraction (IE) methods obtain structured information from unstructured data. One of the challenging IE tasks is Event Extraction (EE) which seeks to derive information about specific incidents and their actors from the text. EE is useful in many domains such as building a knowledge base, information retrieval and summarization. In the past decades, some event ontologies like ACE, CAMEO and ICEWS were developed to define event forms, actors and dimensions of events observed in the text. These event ontologies still have some shortcomings such as covering only a few topics like political events, having inflexible structure in defining argument roles and insufficient gold-standard data. To address these concerns, we propose an event ontology, namely COfEE, that incorporates both expert domain knowledge and a data-driven approach for identifying events from text. COfEE consists of two hierarchy levels (event types and event sub-types) that include new categories relating to environmental issues, cyberspace and criminal activity which need to be monitored instantly. Also, dynamic roles according to each event sub-type are defined to capture various dimensions of events. In a follow-up experiment, the proposed ontology is evaluated on Wikipedia events, and it is shown to be general and comprehensive. Moreover, in order to facilitate the preparation of gold-standard data for event extraction, a language-independent online tool is presented based on COfEE. A gold-standard dataset annotated by 10 human experts is also prepared consisting 24K news articles in Persian language. Finally, we present a supervised method based on deep learning techniques to automatically extract relevant events and corresponding actors.

cs.CL↗

arXiv2020

SimBins: An information-theoretic approach to link prediction in real multiplex networks

The entities of real-world networks are connected via different types of connections (i.e. layers). The task of link prediction in multiplex networks is about finding missing connections based on both intra-layer and inter-layer correlations. Our observations confirm that that in a wide range of real-world multiplex networks, from social to biological and technological, a positive correlation exists between connection probability in one layer and similarity in other layers. Accordingly, a similarity-based automatic general-purpose multiplex link prediction method -- SimBins -- is devised that quantifies the amount of connection uncertainty based on observed inter-layer correlations in a multiplex network. Moreover, SimBins enhances the prediction quality in the target layer by incorporating the effect of link overlap across layers. Applied to various datasets from different domains, SimBins proves to be robust and superior than compared methods in majority of experimented cases in terms of accuracy of link prediction. Furthermore, it is discussed that SimBins imposes minor computational overhead to the base similarity measures making it a potentially fast method, suitable for large-scale multiplex networks.

physics.soc-ph↗

arXiv2019

Link Prediction in Real-World Multiplex Networks via Layer Reconstruction Method

A large body of research on link prediction problem is devoted to finding missing links in single-layer (simplex) networks. The proposed link prediction methods compute a similarity measure between unconnected node pairs based on the observed structure of the network. However, extension of notion of similarity to multiplex networks is a two-fold challenge. The layers of real-world multiplex networks do not have the same organization yet are not of totally different organizations. So, it should be determined that how similar are the layers of a multiplex network. On the other hand, it is needed to be known that how similar layers can contribute in link prediction task on a target layer with missing links. Eigenvectors are known to well reflect the structural features of networks. Therefore, two layers of a multiplex network are similar w.r.t. structural features if they share similar eigenvectors. Experiments show that layers of real-world multiplex networks are similar w.r.t. structural features and the value of similarity is far beyond their randomized counterparts. Furthermore, it is shown that missing links are highly predictable if their addition or removal do not significantly change the network structural features. Otherwise, if the change is significant a similar copy of structural features may come to help. Based on this concept, Layer Reconstruction Method (LRM) finds the best reconstruction of the observed structure of the target layer with structural features of other similar layers. Experiments on real multiplex networks from different disciplines show that this method benefits from information redundancy in the networks and helps the performance of link prediction to stay robust even under high fraction of missing links.

physics.soc-ph↗