Yazar "Kodaz H." seçeneğine göre listele

Listeleniyor 1 - 4 / 4
  • Küçük Resim Yok
    Öğe
    A hybrid method for rating prediction using Linked Data features and text reviews
    (CEUR-WS, 2016) Yumusak S.; Muñoz E.; Minervini P.; Dogdu E.; Kodaz H.
    This paper describes our entry for the Linked Data Mining Challenge 2016, which poses the problem of classifying music albums as 'good' or 'bad' by mining Linked Data. The original labels are assigned according to aggregated critic scores published by the Metacritic website. To this end, the challenge provides datasets that contain the DBpedia reference for music albums. Our approach benefits from Linked Data (LD) and free text to extract meaningful features that help distinguishing between these two classes of music albums. Thus, our features can be summarized as follows: (1) direct object LD features, (2) aggregated count LD features, and (3) textual review features. To build unbiased models, we filtered out those properties somehow related with scores and Metacritic. By using these sets of features, we trained seven models using 10-fold cross-validation to estimate accuracy. We reached the best average accuracy of 87.81% in the training data using a Linear SVM model and all our features, while we reached 90% in the testing data.
  • Küçük Resim Yok
    Öğe
    An improved tree seed algorithm for optimization problems
    (International Association of Computer Science and Information Technology, 2018) Aslan M.; Beskirli M.; Kodaz H.; Kiran M.S.
    Various heuristic algorithms have been proposed in the literature for solving optimization problems. Tree-seed algorithm (TSA) is inspired from relation between trees and seeds that a population-based evolutionary algorithm. when create process of a seed occur in TSA, the position updating of each dimension of the seed is calculated separately. In scope this study, Some changes have been implemented to original TSA. A new operator was added to the position update equation of original TSA when create a seed from tree. This operator is calculated by dynamically according to the dimension of the problem. As the dimension of the problem increases, the value of the this operator decreases. In addition, we determined an upper and a lower bound for the update process of the seed. The Improved Tree Seed Algorithm (ITSA) proposed in this study and the TSA have been tested on some benchmark functions in the literature. As a result, when the experimental results are taken into consideration, it is understood that the proposed algorithm ITSA is obtained more effective results for benchmark functions than TSA. Moreover, it is observed that ITSA found quite successful results compared with TSA for large-scale benchmark problems. © 2018 International Association of Computer Science and Information Technology.
  • Küçük Resim
    Öğe
    A new classification method to diagnosis liver disorders: Supervised Artificial Immune System (AIRS) [Karaci?er rahatsizli?i teşhisinde yeni bir siniflama yöntemi: Danişmali Yapay Ba?işiklik Sistemi (AIRS)]
    (2005) Polat K.; Şahan S.; Kodaz H.; Güneş S.
    Medical diagnosis is very important in medicine. It is necessary to form an efficient and effective computer-based method for decision support in medical analysis. Artificial Immune Systems (AIS), which we can say very new, is an effective and prosperous artificial intelligence area with respect to its problem solving performance. In the beginning, it was formed for helping medical experts to understand the working procedure of immune system in more detail by modeling interactions in immune system. The used medical data was taken from machine learning database of California University in Irvine. In this study, each of data was classified with Artificial Immune Recognition System (AIRS). This application was done by using MATLAB 6.5 programming language. AIRS classification algorithm, which has an important place among classification algorithms in the field of Artificial Immune Systems, has showed an effective and intriguing performance on the problem it was applied. AIRS was previously applied to some medical classification problems including Breast Cancer, Cleverand Heart Disease, Diabetes and it obtained very satisfactory results. So, AIRS proved to be an efficient artificial intelligence technique in medical field. © 2005 IEEE.
  • Küçük Resim Yok
    Öğe
    A short survey of linked data ranking
    (Association for Computing Machinery, Inc, 2014) Yumusak S.; Dogdu E.; Kodaz H.
    Linked data systems are still far from maturity. Hence, the basic principles are still open for discussion. In our study on building a novel linked data search engine, we have surveyed fundamental methods of internet search technologies in the context of linked data crawling, indexing, ranking, and monitoring. The scope of this ranking survey covers linked data related statistical ranking, database ranking, document level ranking, and Web ranking techniques. In order to classify the linked data ranking methods, we identified a number of categories. These categories are ontology ranking, RDF ranking, graph ranking, entity ranking, document/domain ranking. At the end of the survey, we have listed the ranking techniques based on the well-known PageRank algorithm. Copyright 2014 ACM.