Social media responses to news have increasingly gained in importance as they can enhance a consumer's news reading experience, promote information sharing and aid journalists in assessing their readership's response to a story. Given that the number of responses to an online news article may be huge, a common challenge is that of selecting only the most interesting responses for display. This paper addresses this challenge by casting message selection as an optimization problem. We define an objective function which jointly models the messages' utility scores and their entropy. We propose a near-optimal solution to the underlying optimization problem, which leverages the submodularity property of the objective function. Our solution first learns the utility of individual messages in isolation and then produces a diverse selection of interesting messages by maximizing the defined objective function. The intuitions behind our work are that an interesting selection of messages contains diverse, informative, opinionated and popular messages referring to the news article, written mostly by users that have authority on the topic. Our intuitions are embodied by a rich set of content, social and user features capturing the aforementioned aspects. We evaluate our approach through both human and automatic experiments, and demonstrate it outperforms the state of the art. Additionally, we perform an in-depth analysis of the annotated "interesting" responses, shedding light on the subjectivity around the selection process and the perception of interestingness.

No abstract

Članek predstavlja algoritem in implementacijo programa za razpoznavanje imen v slovenskem jeziku s pomočjo strojnega učenja. Nadzorovani pristop na osnovi pogojnih naključnih polj je naučen na označenem korpusu ssj500k. V korpusu, ki je prosto dostopen pod licenco Creative Commons CC-BY-NC-SA, so pri besednih pojavnicah poleg oblikoskladenjskih oznak in lem označena tudi imena organizacij, osebna, zemljepisna ter stvarna imena. Članek predstavlja vpliv na natančnost razpoznavanja ob uporabi oblikoskladenjskih oznak, leksikonov in konjunkcij sosednjih lastnosti. Ena od ugotovitev raziskave je, da so oblikoskladenjske oznake pri razpoznavanju entitet koristne. V kombinaciji z vsemi ostalimi lastnostmi doseže sistem na testni množici 74% natančnost in 72% priklic, pri čemer so najbolje razpoznana osebna imena, sledijo jim zemljepisna ter organizacijska in nazadnje stvarna imena. Novo spoznanje članka je tudi to, da lahko z delitvijo razreda vseh stvarnih imen na organizacije in preostala stvarna imena dosežemo boljše rezultate prepoznavanja tudi pri drugih razredih. Preizkusi na neodvisno označenih korpusi kažejo dobro posplošenost modela za osebna in zemljepisna imena. Programska oprema, narejena v raziskavi, je prosto dostopna pod licenco Apache 2.0 na naslovu http://ailab.ijs.si/~tadej/slner.zip, razvojne različice pa so na voljo na naslovuhttps://github.com/tadejs/slner.

No abstract

This paper presents the linguistic analysis infrastructure developed within the XLike project. The main goal of the implemented tools is to provide a set of functionalities supporting the XLike main objectives: Enabling cross-lingual services for publishers, media monitoring or developing new business intelligence applications. The services cover seven major and minor languages: English, German, Spanish, Chinese, Catalan, Slovenian, and Croatian. These analyzers are provided as web services following a lightweigth SOA architecture approach, and they are publically accessible and shared through META-SHARE. 1