Mission
To provide a Journalists Workbench and Tools for the direct access, efficient processing, in-depth analysis and complete evaluation of content and complex information in monolingual and multilingual written and spoken political and journalistic texts.
Projects
- Imply
Detection of Implicit Information and Connotative features in Written and Spoken Journalistic Texts
- Prag-Graph
Graphic Representation and Pragmatic Evaluation of Interviews, Discussions and Speeches
- Thucydides
Accessing Facts and Diplomacy of the Past – Extracting Information from Ancient “Journalistic” Texts
Head
Christina Alexandris, ΕCI / QJNT, Assoc. Professor, National and Kapodistrian University of Athens
External Researchers:
Dr. Christina Valavani
Vasilios Floros, MSc
Dimitrios Mourouzidis, MSc
Stavros Giannakis, MSc (Opinion Mining and Sentiment Analysis)
External Collaborators:
Mironas Bitinis, MSc
Konstantinos Mylonakis, MSc
Project Information
Imply – Processing Implied Information and Connotative features in Written and Spoken Journalistic Texts
The present approach targets to facilitate the translation, the detailed processing and the correct transfer of opinions, style and overall spirit of written and spoken online journalistic texts. Here, we present the integration of an annotation strategy for written and spoken journalistic texts detecting elements with explicit and implicit connotative features. The proposed annotation strategy is morphologically based and related to a controlled-language-like framework, functioning as a checklist and targeting to address re-occurring problems encountered mainly by “semi-professional” translators, namely journalists, economists and other professionals working with multilingual written and transcribed journalistic texts available from the media and the web. Most of these professionals, usually having an above-average fluency of one or more foreign languages, often lack the necessary exposure to the culture(s) related to the foreign language(s) concerned, especially due to distance or frequent change of location. Thus, essential information presented either in a subtle form or in an indirect way, constituting emotionally and socio-culturally “marked” elements, is often undetected.
The designed user-oriented module is aimed to be integrated in a annotation tool targeting to indicate the largest possible percentage of the points in the texts signalizing “marked” information, alerting the user-translator to evaluate these expressions and, in the case of transcribed spoken journalistic texts, to allow the comparison of “marked” elements with prosodic and paralinguistic features in the respective multimedia files.
Prag-Graph– Graphic Representation and Pragmatic Evaluation of Interviews, Discussions and Speeches
The designed annotation tool targets (1) to provide the User-Journalist with the tracked indications of the topics handled in the interview or discussion and (2) to view the graphic pattern of the discourse structure of the interview or discussion, (3) to evaluate the discourse structure, (4) to allow the User to compare the discourse structure of conversations and interviews with similar topics or the same participants / participant and (5) to indicate the largest possible percentage of the points in the texts signalizing information with implied information and connotative features.
The interface of the annotation tool is designed to (a) to track the “local” topic discussed in a given segment of an interview or discussion or change of “local” topic in an interview or discussion and (b) annotate and highlight all the points possibly containing connotative features information, alerting the User to evaluate the parts of the text containing these expressions.
The designed tool allows the tracking of any change of topic or the same or a similar answer, as well as associations and generalizations related to the same topic.
Incoming texts to be processed constitute transcribed data from journalistic texts. The interactive annotation tool is designed to operate with most commercial transcription tools, some of which are available online. The designed tool may also be adapted to downloaded written texts from the internet (blog).
Thucydides – Accessing Facts and Diplomacy of the Past – Extracting Information from Ancient “Journalistic” Texts
For the International Public, ancient historical and “journalistic” texts, such the “Peloponnesian War” of the Ancient Greek historian Thucydides, may allow an insight for the understanding of current international and national political affairs and international political and economic relations. The present approach targets to facilitate the accessibility of such texts for non-experts in the International Public, especially journalists, translators and students. Specifically, the basic issue to be addressed here is the possibility to access complex information in the Ancient Text related to diplomacy and to compare it to passages from online journalistic texts (1) and to directly find out respective passages in the original texts along with a translation in English (2) as well as a second type of translation containing structures close to the original text, minimizing language-specific interference and parameters of translations (3). The latter possibility (3) provides a closer look to the content and structure of the original text and is less dependent on language-specific parameters interfering in the English translation.
The present approach concerns the integration of expert knowledge within a System-controlled framework for the detection of information concerning diplomacy, especially cause and result relations contained in the online Ancient Text. The module presented here is designed to make use of already-existing tools and mechanisms, the construction of a database and interface with low computational cost, combined with expert knowledge and sublanguage – specific parameters. For the handling of topics related to complex information such as “Diplomacy”, expert knowledge and sublanguage – specific parameters are put to use to constitute a framework replacing conventional information extraction methods and statistically-based approaches.
For more click here