The paper presents a new corpus for fake news detection in the Urdu language along with the baseline classification and its evaluation. With the escalating use of the Internet worldwide and substantially increasing impact produced by the availability of ambiguous information, the challenge to quickly identify fake news in digital media in various languages becomes more acute. We provide a manually assembled and verified dataset containing 900 news articles, 500 annotated as real and 400, as fake, allowing the investigation of automated fake news detection approaches in Urdu. The news articles in the truthful subset come from legitimate news sources, and their validity has been manually verified. In the fake subset, the known difficulty of finding fake news was solved by hiring professional journalists native in Urdu who were instructed to intentionally write deceptive news articles. The dataset contains 5 different topics: (i) Business, (ii) Health, (iii) Showbiz, (iv) Sports, and (v) Technology. To establish our Urdu dataset as a benchmark, we performed baseline classification. We crafted a variety of text representation feature sets including word n-grams, character n-grams, functional word n-grams, and their combinations. After applying a variety of feature weighting schemes, we ran a series of classifiers on the train-test split. The results show sizable performance gains by AdaBoost classifier with 0.87 F1Fake and 0.90 F1Real. We provide the results evaluated against different metrics for a convenient comparison of future research. The dataset is publicly available for research purposes.

Automatic threatening language detection is an important task and most of the existing studies relied on English. However, threatening language detection in poor-resource language remains briefly addressed. In this paper, we introduce a new publicly available dataset for threatening language detection in Urdu tweets to fill the scientific gap, particularly, in the Urdu language. The proposed dataset contains 3,564 tweets manually annotated by human experts with two labels: threatening and non-threatening. The threatening tweets are further classified into two classes: threatening to an individual person or threatening to a group. This research follows a twostep approach: (i) classify a given tweet as threatening or non-threatening and (ii) classify whether a threatening tweet is used to threaten an individual or a group. We compare three forms of text representation: two count-based, where the text is represented using either character n-gram counts or word n-gram counts as feature vectors and the third text representation is based on fastText pre-trained word embeddings for Urdu. We perform several experiments using machine learning and deep learning classifiers and our study shows that MLP classifier with the combination of word n-gram features outperformed other classifiers in detecting threatening tweets. Whereas, SVM using fastText pre-trained word embedding obtained the best results for the target identification task.

This study reports the second shared task named as UrduFake@Fire2021 on identifying fake news detection in Urdu language. This is a binary classification problem in which the task is to classify a given news article into two classes: (i) real news, or (ii) fake news. In this shared task, 34 teams from 7 different countries (China, Egypt, Israel, India, Mexico, Pakistan, and UAE) registered to participate in the shared task, 18 teams submitted their experimental results and 11 teams submitted their technical reports. The proposed systems were based on various count-based features and used different classifiers as well as neural network architectures. The stochastic gradient descent (SGD) algorithm outperformed other classifiers and achieved 0.679 F-score.

This paper gives the overview of the first shared task at FIRE 2020 on fake news detection in the Urdu language. This is a binary classification task in which the goal is to identify fake news using a dataset composed of 900 annotated news articles for training and 400 news articles for testing. The dataset contains news in five domains: (i) Health, (ii) Sports, (iii) Showbiz, (iv) Technology, and (v) Business. 42 teams from 6 different countries (India, China, Egypt, Germany, Pakistan, and the UK) registered for the task. 9 teams submitted their experimental results. The participants used various machine learning methods ranging from feature-based traditional machine learning to neural network techniques. The best performing system achieved an F-score value of 0.90, showing that the BERT-based approach outperforms other machine learning classifiers.

Open Information Extraction (Open IE) serves for the analysis of vast amounts of texts by extraction of assertions, or relations, in the form of tuples argument 1; relation; argument 2. Various approaches to Open IE have been designed to perform in a fast, unsupervised manner. All of them require language specific information for their implementation. In this work, we introduce an approach to Open IE based on syntactic constraints over POS tag sequences targeted at Spanish language. We describe the rules specific for Spanish language constructions and their implementation in EXTRHECH, an Open IE system for Spanish. We also discuss language-specific issues of implementation. We compare EXTRHECH's performance with that of REVERB, a similar Open IE system for English, on a parallel dataset and show that these systems perform at a very similar level. We also compare EXTRHECH's performance on a dataset of grammatically correct sentences against its performance on a dataset of random texts extracted from the Web, drastically different in their quality from the first dataset. The latter experiment shows robustness of EXTRHECH on texts from the Web.