Sentinel: A Codesigned Platform for Semantic Enrichment of Social Media Streams

Abstract

We introduce the Sentinel platform that supports semantic enrichment of streamed social media data for the purposes of situational understanding. The platform is the result of a codesign effort between computing and social scientists, iteratively developed through a series of pilot studies. The platform is founded upon a knowledge-based approach, in which input streams (channels) are characterized by spatial and terminological parameters, collected media is preprocessed to identify significant terms (signals), and data are tagged (framed) in relation to an ontology. Interpretation of processed media is framed in terms of the 5W framework (who, what, when, where, and why). The platform is designed to be open to the incorporation of new processing modules, building on the knowledge-based elements (channels, signals, and framing ontology) and accessible via a set of user-facing apps. We present the conceptual architecture for the platform, discuss the design and implementation challenges of the underlying streamprocessing system, and present a number of apps developed in the context of the pilot studies, highlighting the strengths and importance of the codesign approach and indicating promising areas for future research.

Techniques of disinformation: Constructing and communicating “soft facts” after terrorism

“…A total of just over 30 million data points were collated from across multiple social media platforms, with a particular focus upon Twitter, utilizing the Sentinel platform. Sentinel comprises a suite of data collection and analysis “apps” and algorithms, with similar collection and processing functionality to many commercial packages (Preece et al, ). However, whereas the latter are “black boxed” (Pasquale, ), Sentinel is “glass boxed,” enabling researchers to investigate how particular decisions and choices in terms of data collection, processing and analysis, structure and shape resultant data flows.…”

Hyperlink Induced Topic Search Model Together with Automatic Feature Review for Smartphone Applications

“…The university is now applying affiliation guidelines to individual recommendations. Liu et al [19] investigated how people with similar tastes and desires would receive tips on Weibo. They suggested a data mining-based suggestion algorithm.…”

“…Furthermore, if the weight of each object is not taken into account, a vast number of rules will be detected and the advice for applications is insignificant. In order to evaluate a set of web pages appropriate for one subject and find the "authoritative" ones in the topic, the HITS algorithm was initially implemented in [20] and [19]. On these web pages, it conducts connect analytics to classify them as two measures: Hub Meaning and Authority.…”

Normalisation et domestication de la désinformation numérique : les opérations informationnelles d’interférence et d’influence de l’extrême droite et de l’État russe en Europe

“…Les données récoltées ont été organisées par plateforme, avec une attention particulière pour Twitter, en utilisant le logiciel Sentinel. Il comprend un logiciel de collecte des données, d'algorithmes et d'applications à des fins d'analyse, ainsi que des fonctionnalités de collecte et de traitement similaires aux logiciels commerciaux [Preece et al, 2018]. Mais, si ces suites commerciales sont des « boîtes noires » [Pasquale, 2015], Sentinel est une « boîte en verre » qui permet aux chercheurs d'étudier comment une décision ou un choix particulier fait dans la collecte, le traitement et l'analyse des données structure et modifie en conséquence les flux de données entrant.…”