In this paper, we investigate the effects of using subword information in representation learning. We argue that using syntactic subword units effects the quality of the word representations positively. We introduce a morpheme-based model and compare it against to word-based, characterbased, and character n-gram level models. Our model takes a list of candidate segmentations of a word and learns the representation of the word based on different segmentations that are weighted by an attention mechanism. We performed experiments on Turkish as a morphologically rich language and English with a comparably poorer morphology. The results show that morpheme-based models are better at learning word representations of morphologically complex languages compared to character-based and character ngram level models since the morphemes help to incorporate more syntactic knowledge in learning, that makes morphemebased models better at syntactic tasks.

We present a fully unsupervised method for morphological segmentation. Unlike many morphological segmentation systems, our method is based on semantic features rather than orthographic features. In order to capture word meanings, word embeddings are obtained from a two-level neural network [11]. We compute the semantic similarity between words using the neural word embeddings, which forms our baseline segmentation model. We model morphotactics with a bigram language model based on maximum likelihood estimates by using the initial segmentations from the baseline. Results show that using semantic features helps to improve morphological segmentation especially in agglutinating languages like Turkish. Our method shows competitive performance compared to other unsupervised morphological segmentation systems.

No abstract

Since mobile applications make our lives easier, there is a large number of mobile applications customized for our needs in the application markets. While the application markets provide us a platform for downloading applications, it is also used by malware developers in order to distribute their malicious applications. In Android, permissions are used to prevent users from installing applications that might violate the users' privacy by raising their awareness. From the privacy and security point of view, if the functionality of applications is given in suffcient detail in their descriptions, then the requirement of requested permissions could be well understood. This is defned as description-topermission fdelity in the literature. In this study, we propose two novel models that address the inconsistencies between the application descriptions and the requested permissions. The proposed models are based on the current state-of-art neural architectures called attention mechanisms. Here, we aim to fnd the permission statement words or sentences in app descriptions by using the attention mechanism along with recurrent neural networks. The lack of such permission statements in application descriptions creates a suspicion. Hence, the proposed approach could assist in static analysis techniques in order to fnd suspicious apps and to prioritize apps for more resource intensive analysis techniques. The experimental results show that the proposed approach achieves high accuracy.

In this article, we investigate using deep neural networks with different word representation techniques for named entity recognition (NER) on Turkish noisy text. We argue that valuable latent features for NER can, in fact, be learned without using any hand-crafted features and/or domain-specific resources such as gazetteers and lexicons. In this regard, we utilize character-level, character n-gram-level, morpheme-level, and orthographic character-level word representations. Since noisy data with NER annotation are scarce for Turkish, we introduce a transfer learning model in order to learn infrequent entity types as an extension to the Bi-LSTM-CRF architecture by incorporating an additional conditional random field (CRF) layer that is trained on a larger (but formal) text and a noisy text simultaneously. This allows us to learn from both formal and informal/noisy text, thus improving the performance of our model further for rarely seen entity types. We experimented on Turkish as a morphologically rich language and English as a relatively morphologically poor language. We obtained an entity-level F1 score of 67.39% on Turkish noisy data and 45.30% on English noisy data, which outperforms the current state-of-art models on noisy text. The English scores are lower compared to Turkish scores because of the intense sparsity in the data introduced by the user writing styles. The results prove that using subword information significantly contributes to learning latent features for morphologically rich languages.

No abstract

This paper is a survey of methods and algorithms for unsupervised learning of morphology. We provide a description of the methods and algorithms used for morphological segmentation from a computational linguistics point of view. We survey morphological segmentation methods covering methods based on MDL (minimum description length), MLE (maximum likelihood estimation), MAP (maximum a posteriori), parametric and non-parametric Bayesian approaches. A review of the evaluation schemes for unsupervised morphological segmentation is also provided along with a summary of evaluation results on the Morpho Challenge evaluations.

Özet-Türkçe, morfem adı verilen birimlerin art arda eklenmesiyle sözcüklerin oluşturulduğu sondan eklemeli bir dildir. Sözcüklerin farklı parçaların birleştirilmesiyle oluşturulması makine tercümesi, duygu analizi ve bilgi çıkarımı gibi birçok doğal dil işleme uygulamasında seyreklik problemine yol açmaktadır çünkü sözcüğün her farklı formu farklı bir sözcük gibi algılanmaktadır. Bu makalede, sözcüklerin yapım ve çekim eklerinden arındırılarak köklerinin otomatik olarak bulunabilmesi için bir yöntem öneriyoruz. Kullandığımız yöntem tekrarlayan sinir ağları kullanarak oluşturulan kodlayıcı-kod çözücü yaklaşımına dayanmaktadır. Verilen herhangi bir sözcük, oluşturduğumuz sinir ağı yapısı ile öncelikle kodlanmakta, ardından kodu çözülerek köküne ulaşılabilmektedir. Bu yöntem şimdiye kadar etiketleme veya makine tercümesi gibi problemlerde kullanılmıştır. Diğer Türkçe kök bulma modelleriyle karşılaştırıldığında sonuçların oldukça iyi olduğu gözlenmiştir. Diğer modellerde olduğu gibi, herhangi bir kural kümesi elle tanımlanmadan, sadece sözcük ve kök ikililerinden oluşan bir eğitim veri kümesi kullanılarak kök bulma işlemi önerdiğimiz bu model ile gerçekleştirilebilmektedir.