Browsing by Author "Kişla T."

Now showing 1 - 2 of 2

Citation - Scopus: 1
Combining Machine Translation and Text Similarity Metrics To Identify Paraphrases in Turkish
(Institute of Electrical and Electronics Engineers Inc., 2018) Soleymanzadeh K.; Karaoğlan B.; Metin S.K.; Kişla T.
Paraphrase identification (PI) is to recognize whether given two sentences are restatements of each other or not. In our study we propose an approach that exploits machine translation and text similarity metrics as features for PI. Machine learning algorithms like Support Vector Machine (SVM) with three different kernels, C4.5 Decision tree and Multinomial Naïve Bayes (NB) are trained with these features. We evaluated our system on Parder, Turkish paraphrase corpus. The experimental results show that the proposed approach offers state-of-the-art results. © 2018 IEEE.
Verb Detection in Turkish Using Logistic Regression Analysis
(2011) Kumova Metin, Senem; Kişla T.; Karaog?lan B.
In this paper, we investigated the features which discriminates verbs in Turkish. Though, words in Turkish can be classified in eight different linguistic categories (noun, verb, adjective, pronoun, adverb, postposition, conjunction and interjection), they can be discriminated in two fundamental groups as nouns and verbs. We have utilized logistic regression method to determine and compare the features based on sentence and word properties considering that all linguistic categories except verbs can be merged into the to the noun originated group. The strength of both categorical (such as inclusion of capital letters, apostrophes) and numerical features (such as position in sentence) in verb discrimination are examined and results are presented. We believe that this study may contribute to the time consuming rule based or probabilistic POS tagging applications since logistic regression analysis gives an immediate foresight of verbs. © 2011 Praise Worthy Prize S.r.l.