Clustering Analysis for Vasculitic Diseases

Yildirim P., Ceken C., ÇEKEN K., Tolun M. R.

2nd International Conference on Networked Digital Technologies, Prague, Çek Cumhuriyeti, 7 - 09 Temmuz 2010, cilt.88, ss.36-38 identifier identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Cilt numarası: 88
  • Doi Numarası: 10.1007/978-3-642-14306-9_5
  • Basıldığı Şehir: Prague
  • Basıldığı Ülke: Çek Cumhuriyeti
  • Sayfa Sayıları: ss.36-38
  • Anahtar Kelimeler: Biomedical text mining, data mining, clustering analysis, vasculitic diseases, TEXT-MINING SYSTEM
  • Akdeniz Üniversitesi Adresli: Evet

Özet

We introduce knowledge discovery for vasculitic diseases in this paper. Vasculitic diseases affect some organs and tissues and diagnosing can be quite difficult. Biomedical literature can contain hidden and useful knowledge for biomedical research and we develop a study based on co-occurrence analysis by using the articles in MEDLINE which is a widely used database. The mostly seen vasculitic diseases are selected to explore hidden patterns. We select PolySearch system as a web based biomedical text mining tool to find organs and tissues in the articles and create two separate datasets with their frequencies for each disease. After forming these datasets, we apply hierarchical clustering analysis to find similarities between the diseases. Clustering analysis reveals some similarities between diseases. We think that the results of clustered diseases positively affect on the medical research of vasculitic diseases especially during the diagnosis and certain similarities can provide different views to medical specialists.