Hacer Karacan

Is attention drawn to changes in familiar scenes

In this study, we examined the mechanisms that control attention in natural scenes. We asked whether familiarity with the environment makes subjects more sensitive to changes or novel events in the scene. Previous investigation of this issue has been based on viewing 2-D images of simple objects or of natural scenes, a situation that does not accurately reflect the challenges of natural vision. We found that familiarity with the environment significantly increased the time spent fixating regions in the scene where a change had occurred. Together with previous work (Brockmole & Henderson, 2005a, 2005b), our results support the hypothesis that we learn the structure of natural scenes over time, and that attention is attracted by deviations from the stored scene representation. Such a mechanism would allow deployment of attention to objects or events that were not explicitly on the current cognitive agenda.

Ontology creation for an educational center

The introduction of the Semantic Web in 2001 lead the way to many new ideas and technologies in computing. Data handling and sharing gained a semantic level rather than structural, facilitating the development of innovative and more helpful end-user applications. Semantics of Web information is formally defined in ontologies, making it possible for machines to interpret data content more effectively. In this study, we aimed to apply this new technology and create a useful application for an educational center. This paper presents our work, issues we encountered, our approach to resolving those issues, as well as our findings and results.

VERİ MADENCİLİĞİNDE NESNE YÖNELİMLİ BİRLEŞTİRİCİ HİYERARŞİK KÜMELEME MODELİ

Veri madenciligi alaninda farkli kullanim amacina sahip cok sayida teknik mevcuttur. Son yillarda siklikla kullanilan tekniklerden biri de birlestirici hiyerarsik kumeleme analizidir. Bu calismada nesne yonelimli birlestirici hiyerarsik kumeleme modeli gelistirilmistir. Gelistirilen modelin interaktifligi sayesinde kullanicilarin sistem arayuzunu rahatlikla kullanabilmesi ve nesne yonelimli programlama sayesinde de kullanici taleplerinin hizli bir sekilde karsilanmasi saglanmistir. Gelistirilen model bu sayede optimum performansa sahip olmustur. Modelin test asamasinda, Turkiye’deki 44 ile ait aylik ortalama ruzgâr hizi verileri hiyerarsik olarak kumelenmistir. Kumeleme sonuclari Microsoft Visual Studio.NET platformunda olusturulan arayuzde metinsel formatta sunulurken; Matlab’da ise dendrogramlar araciligiyla gorsellestirilmistir. Kumeleme sonuclari sayesinde, onceden gizli olan fakat model sayesinde ortaya cikarilan pek cok cikarim elde edilmis ve model analizi etkin bir sekilde yapilmistir.