Stalker, a Multilingual Text Mining Search Engine for Open Source Intelligence

Abstract

Open Source Intelligence (OSINT) is an intelligence gathering discipline that involves collecting information from open sources and analyzing it to produce usable intelligence. The international Intelligence Communities have seen open sources grow increasingly easier and cheaper to acquire in recent years. But up to 80% of electronic data is textual and most valuable information is often hidden and encoded in pages which are neither structured, nor classified. The process of accessing all these raw data, heterogeneous in terms of source and language, and transforming them into information is therefore strongly linked to automatic textual analysis and synthesis, which are greatly related to the ability to master the problems of multilinguality. This paper describes a content enabling system that provides deep semantic search and information access to large quantities of distributed multimedia data for both experts and general public. STALKER provides with a language independent search and dynamic classification features for a broad range of data collected from several sources in a number of culturally diverse languages.

Discovering Hidden Concepts in Predictive Models for Texts' Polarization

“…Such documents have been read and classified by the company, according to its communication assets and brand mission. Then, following the I step of our protocol we computed the lemmas (key-words) of our corpus employing SyN Semantic Center (Neri & Raffaelli, 2004;Neri & Pettoni, 2009), a complex system of text intelligence analytical-linguistic, produced by Synthema, (an Italian company of Human Language Technology). SyN Semantic Center is an advanced technology platform that allows running linguistic and semantic analysis of any piece of information, such as documents, web pages, discussion groups, forums, chats, e-mails, databases, scientific and technical publications.…”

“…Hence, the ambitious aim of our paper is to deal with both issues, in order to provide an efficient and robust supervised classification rule, which is able, not only to discriminate sentiments extracted from the customer opinions, but also to identify texts significantly characterizing different levels of liking or dislike rewferred to a brand/product. Therefore, following Liberati & Camillo (2014) approach, we employed an intelligent semantic lemmatization (Neri & Raffaelli, 2004;Neri & Pettoni, 2009) for analyzing unstructured data, coming from different web sources: forum, blogs and social networks. Then, we estimated a probabilistic classifier based on kernel machines to get the polarization rule, but any other probabilistic classifier with comparable performance could be applied.…”

OSIA: Open Source Intelligence Analysis System Based on Cloud Computing and Domestic Platform

“…However, almost all intelligence analysis systems are based on Windows. There are many media monitor applications, such as Europe Media Monitor [5], NRTIM(Near Real Time Information Mining in Multilingual News) [6], STALKER [7], and many business intelligence software, like Actuate, JasperSoft, OpenI, Palo, Pentaho and SpagoBI [8]. It is really urgent that an intelligence analysis system based on domestic platform operating system is required.…”