Leona Van Vaerenbergh

Predicting age and gender in online social networks

A common characteristic of communication on online social networks is that it happens via short messages, often using non-standard language variations. These characteristics make this type of text a challenging text genre for natural language processing. Moreover, in these digital communities it is easy to provide a false name, age, gender and location in order to hide ones true identity, providing criminals such as pedophiles with new possibilities to groom their victims. It would therefore be useful if user profiles can be checked on the basis of text analysis, and false profiles flagged for monitoring. This paper presents an exploratory study in which we apply a text categorization approach for the prediction of age and gender on a corpus of chat texts, which we collected from the Belgian social networking site Netlog. We examine which types of features are most informative for a reliable prediction of age and gender on this difficult text type and perform experiments with different data set sizes in order to acquire more insight into the minimum data size requirements for this task.

ABOP, the automatic patient information leaflet optimizer: evaluation of a tool in development.

OBJECTIVE We developed a semi-automated leaflet optimizer (ABOP) to improve the readability of Dutch patient information leaflets (PILs). Our aim was to assess whether or not revisions made with ABOP have an effect on traceability and comprehension of PIL information. METHODS Two leaflets (one for oxazepam and one for tetracycline) were revised with the ABOP tool and both the original versions and ABOP versions were tested for readability, according to the EC guideline. RESULTS The ABOP version for oxazepam (p<.0001) scored significantly better than its original counterpart in readability tests. Although the results showed that the ABOP version of the tetracycline antibiotic outperformed the original version the difference in scores was not statistically significant. Terminology improvements had the greatest effect on information retrieval and comprehension. Improvements to instructions and warnings did not affect retrieval and comprehension scores. Overall improvements did not ensure full compliance with EC requirements, but results are not far off. CONCLUSION This study shows that even though ABOP does not solve all text quality issues, it does largely prepare a PIL for readability testing. PRACTICE IMPLICATIONS Using ABOP as a revision tool for PILs can optimize PIL quality and hence reduce readability testing time and costs.

Text creation in a multilingual institutional setting: The translator as part of a cooperative system

Text creation is influenced by situational factors, like the context in which the text is written. In a multilingual institution, the context is especially important, because it is controlled by rules and guidelines. This paper presents a case study conducted in the European Commission and the Directorate General for Translation. The text creation and translation process were reconstructed starting from the published version in the Official Journal of the European Union back to the initial idea. The paper focuses on the impact that external processes have on internal processes. We found that text editing and translation go hand in hand and that translators are part of a cooperative system in which their input is of importance for the source text. Moreover, the impact of the institution is significant, since it determines not only the work processes, but also the linguistic form, technical medium and technical content.

Text creation in a multilingual institutional setting

Text creation is influenced by situational factors, like the context in which the text is written. In a multilingual institution, the context is especially important, because it is controlled by rules and guidelines. This paper presents a case study conducted in the European Commission and the Directorate General for Translation. The text creation and translation process were reconstructed starting from the published version in the Official Journal of the European Union back to the initial idea. The paper focuses on the impact that external processes have on internal processes. We found that text editing and translation go hand in hand and that translators are part of a cooperative system in which their input is of importance for the source text. Moreover, the impact of the institution is significant, since it determines not only the work processes, but also the linguistic form, technical medium and technical content.