PubMed bilgi geri getirim sistemini kullanarak makalelere otomatik anahtar kelime atama

Assigning keywords to research articles is a very important process. These keywords should describe the article properly. Performing this process manually is difficult and may cause an improper description of the article. Therefore, in this thesis, we designed and developed an automatic keyword suggestion system for research articles. In the application we developed two different corpus-based methods by utilizing information retrieval systems using Medline dataset in PubMed. First, proposed keyword suggestion system accepts abstract of the research article as a query to information retrieval system. Next, the information retrieval system returns a list of articles to the given query in ranked order of similarity. Then, we selected a set of documents from this list using two different methods: k-NN and t NN representing the first k documents and documents whose similarity is greater than threshold value of t, respectively. To evaluate our proposed systems, we conducted a set of experiments using randomly chosen a thousand of articles, and provide a comparison of our system results with authors' keywords. The results we obtained showed that our system suggest keywords more than 42% match in terms of F-measure.