Marwa Graja

Orthographic transcription for spoken tunisian arabic

Transcribing spoken Arabic dialects is an important task for building speech corpora. Therefore, it is necessary to follow a definite orthography and a definite annotation to transcribe speech data. In this paper, we present OTTA, Orthographic Transcription for Tunisian Arabic. This convention proposes the use of some rules based on the standard Arabic transcription conventions and we define a set of conventions which preserve the particularities of Tunisian dialect.

Discriminative framework for spoken tunisian dialect understanding

In this paper, we propose to evaluate the performance of a discriminative model to semantically label spoken Tunisian dialect turns which are not segmented into utterances. We evaluate discriminative algorithm based on Conditional Random Fields (CRF). We check the performance of the CRF model to concept labeling on raw data in Tunisian dialect which are not analyzed in advance. We compared its performance with different types of preprocessing data until arriving to well treated data. CRF model showed the ability to ameliorate the accuracy of labeling task for spoken language understanding of not segmented and not treated speech in Tunisian dialect.

Statistical framework with knowledge base integration for robust speech understanding of the Tunisian dialect

In this paper, we propose a hybrid method for the spoken Tunisian dialect understanding within a limited task. This method couples a discriminative statistical method with a domain ontology. The statistical method is based on conditional random field (CRF) models learned from a little size corpus to perform conceptual labeling task. These models are able to detect the semantic dependency between words. However, the domain ontology is used to add prior knowledge about the task. Our experiments are based on a real spoken Tunisian dialect corpus. The obtained results show that the proposed method is able to improve the performance of CRF models for speech understanding by the integration of the domain ontology. Our method can be exploited for under-resourced languages and Arabic dialects to overcome the lack of linguistic resources .

Towards Understanding Spoken Tunisian Dialect

This paper presents a method for semantic interpretation designed for Tunisian dialect. Our method is based on lexical semantics to overcome the lack of resources for the studied dialect. This method is Ontology-based which allows exploiting the ontological concepts for semantic annotation and ontological relations for interpretation. This combination reduces inaccuracies and increases the rate of comprehension. This paper also details the process of building the Ontology used for annotation and interpretation of Tunisian dialect utterances in the context of speech understanding in dialogue systems.

Arabic Fine-Grained Opinion Categorization Using Discriminative Machine Learning Technique

This paper presents an approach of fine-grained opinion categorization in Arabic news articles. This approach is based on lexical semantic analysis. We propose to categorize every opinion expression using a proposed typology of four top-level semantic categories: reporting, judgment, advice and sentiment. Each word or opinion expression will be annotated with a semantic representation which takes in consideration specificities of Arabic language. To the best of our knowledge, there is no annotated Arabic opinion corpus with the proposed semantic representation. The task of categorization is considered as a classification problem. So, we use a Conditional Random Fields (CRF) as a discriminative model that we consider as a good contribution, because of the lack of similar fine-grained opinion categorization performed with CRF. The obtained results show that the integration of CRF models is important for opinion classification of the Arabic language.

Opinion Target Extraction from Arabic News Articles Using shallow Features

Target identification is one of the important tasks related to opinion mining. Indeed, there are few works in this field that deals with Arabic Language because of the lack of annotated corpora. In this paper, we propose to investigate the problem of opinion target identification from Arabic news articles using Conditional Random Fields (CRF) as discriminative framework. Opinion target recognition task consists in determining terms forming the target span. To the best of our knowledge, there is no similar work done in this field for Arabic language and especially for news articles. Experiments show that we can perform excellent results with consideration of semantic correlation between words and without relying on deep syntactic features. Our proposed method identifies opinion target with 95% F-measure, for a given opinion word using bi-gram feature, words in context and other features.

Towards Arabic semantic opinion mining: identifying opinion, polarity and intensity

Arabic opinion mining is a challenging task because Arabic is morphologically and semantically rich language. In this paper, we are interested in analyzing opinions in Arabic news articles. We propose to use a machine learning technique to classify opinions or sentiments at the expression level. Our approach involves determining the semantic category of the expression. It also includes the classification of the opinion expression into positive or negative and the classification of its intensity into high, medium and low. Our method relies on wide range of features which are used in the literature like n-grams, morphological, stylistic features, etc. In addition, we propose new features inspired from contextual, semantic information and others specific for Arabic language. In the same context, we try to have a good contribution in opinion mining in Arabic by proposing to use Conditional Random Fields as a discriminative model. We carry out many experiments by combining at the same time different set of features to find the best combination that yield the best results. We evaluate our method at the expression level using a corpus of Arabic news articles. Our method achieves a good result that reaches 84.93% for contextual polarity classification and 87.54% for semantic opinion expression categorization.