Abstract In this current exercise, we took the FIRE dataset and evaluated the Information Retrieval task using a framework called as Task 1. The Artificial Intelligence Framework for Legal Assistance is used to evaluate the results with the cases for a given situation. In this work the topic terms are extracted for a specific condition and queries are assessed for identification... In this current exercise, we took the FIRE dataset and evaluated the Information Retrieval task using a framework called as Task 1. The Artificial Intelligence Framework for Legal Assistance is used to evaluate the results with the cases for a given situation. In this work the topic terms are extracted for a specific condition and queries are assessed for identification of earlier similar cases. The results are posted in this paper. Read More Read Less

Abstract In this paper, we have worked on irony detection in the Arabic language, a task which is organized by FIRE 2019. The tweets have been preprocessed and tokenized to extract the frequency-based, emotion-based features. These features are used to irony identification using the voted classier. The F-score of our proposed approach is 0.807 and the topranking developed method having F-score... In this paper, we have worked on irony detection in the Arabic language, a task which is organized by FIRE 2019. The tweets have been preprocessed and tokenized to extract the frequency-based, emotion-based features. These features are used to irony identification using the voted classier. The F-score of our proposed approach is 0.807 and the topranking developed method having F-score of .037, so the difference between F-score makes our approach better. Read More Read Less

Abstract In this paper, we describe and show the results of our 3 systems submitted for the Irony Detection in Arabic Tweets Shared Task at the Forum for Information Retrieval (FIRE 2019). We employ ensemble learning for this task through 3 different types of ensemble models, namely classical, deep and hybrid (that combines both). We extract types of features from the... In this paper, we describe and show the results of our 3 systems submitted for the Irony Detection in Arabic Tweets Shared Task at the Forum for Information Retrieval (FIRE 2019). We employ ensemble learning for this task through 3 different types of ensemble models, namely classical, deep and hybrid (that combines both). We extract types of features from the tweets including TF-IDF word n-gram features, topic modeling features, bag-of-words and sentiment features. Our submitted systems scored the top 3 places with our best system achieving 84.4 F1 points on the test set. Read More Read Less

Abstract Recent public calls for the development of explainable and variable Artificial Intelligence (AI) led to a growing interest in formal verification and repair of machine-learned models. Despite the impressive progress that the learning community has made, models such as deep neural networks remain vulnerable to adversarial attacks, and their sheer size represents a major obstacle to formal analysis and implementation.... Recent public calls for the development of explainable and variable Artificial Intelligence (AI) led to a growing interest in formal verification and repair of machine-learned models. Despite the impressive progress that the learning community has made, models such as deep neural networks remain vulnerable to adversarial attacks, and their sheer size represents a major obstacle to formal analysis and implementation. In this paper, we present our current efforts to tackle repair of deep convolutional neural networks using ideas borrowed from Transfer Learning. Using results obtained on popular MNIST and CIFAR10 datasets, we show that models of deep convolutional neural networks can be transformed into simpler ones preserving their accuracy, and we discuss how formal repair through convex programming techniques could benefit from this process. Read More Read Less