No abstract

Misinformation is considered a threat to our democratic values and principles. The spread of such content on social media polarizes society and undermines public discourse by distorting public perceptions and generating social unrest while lacking the rigor of traditional journalism. Transformers and transfer learning proved to be state-of-the-art methods for multiple well-known natural language processing tasks. In this paper, we propose MisRoBÆRTa, a novel transformer-based deep neural ensemble architecture for misinformation detection. MisRoBÆRTa takes advantage of two state-of-the art transformers, i.e., BART and RoBERTa, to improve the performance of discriminating between real news and different types of fake news. We also benchmarked and evaluated the performances of multiple transformers on the task of misinformation detection. For training and testing, we used a large real-world news articles dataset (i.e., 100,000 records) labeled with 10 classes, thus addressing two shortcomings in the current research: (1) increasing the size of the dataset from small to large, and (2) moving the focus of fake news detection from binary classification to multi-class classification. For this dataset, we manually verified the content of the news articles to ensure that they were correctly labeled. The experimental results show that the accuracy of transformers on the misinformation detection problem was significantly influenced by the method employed to learn the context, dataset size, and vocabulary dimension. We observe empirically that the best accuracy performance among the classification models that use only one transformer is obtained by BART, while DistilRoBERTa obtains the best accuracy in the least amount of time required for fine-tuning and training. However, the proposed MisRoBÆRTa outperforms the other transformer models in the task of misinformation detection. To arrive at this conclusion, we performed ample ablation and sensitivity testing with MisRoBÆRTa on two datasets.

Due to the exponential growth of the Internet of Things networks and the massive amount of time series data collected from these networks, it is essential to apply efficient methods for Big Data analysis in order to extract meaningful information and statistics. Anomaly detection is an important part of time series analysis, improving the quality of further analysis, such as prediction and forecasting. Thus, detecting sudden change points with normal behavior and using them to discriminate between abnormal behavior, i.e., outliers, is a crucial step used to minimize the false positive rate and to build accurate machine learning models for prediction and forecasting. In this paper, we propose a rule-based decision system that enhances anomaly detection in multivariate time series using change point detection. Our architecture uses a pipeline that automatically manages to detect real anomalies and remove the false positives introduced by change points. We employ both traditional and deep learning unsupervised algorithms, in total, five anomaly detection and five change point detection algorithms. Additionally, we propose a new confidence metric based on the support for a time series point to be an anomaly and the support for the same point to be a change point. In our experiments, we use a large real-world dataset containing multivariate time series about water consumption collected from smart meters. As an evaluation metric, we use Mean Absolute Error (MAE). The low MAE values show that the algorithms accurately determine anomalies and change points. The experimental results strengthen our assumption that anomaly detection can be improved by determining and removing change points as well as validates the correctness of our proposed rules in real-world scenarios. Furthermore, the proposed rule-based decision support systems enable users to make informed decisions regarding the status of the water distribution network and perform effectively predictive and proactive maintenance.

New mass media paradigms for information distribution have emerged with the digital age. With new digital-enabled mass media, the communication process is centered around the user, while multimedia content is the new identity of news. Thus, the media landscape has shifted from mass media to personalized social media. While this progress brings advantages, it also carries the risk of being detrimental to society through the emergence of misinformation (false or inaccurate information) and disinformation (intentionally spreading misinformation) in the form of fake news. Fake news is a tool used to manipulate public opinion on particular topics, distort public perceptions, and generate social unrest while lacking the rigor of traditional journalism. Driven by this current and real-world problem, in this paper, we train multiple Deep Learning architectures for multi-class classification and compare their performance in detecting the veracity of the news articles. To achieve accurate models in detecting misinformation, we employ a large dataset containing 100 000 news articles labeled with ten classes (one with real news and the rest with different types of fake news). We use two preprocessing techniques, i.e., one simple and another very aggressive, to clean the dataset. We also employ three word embeddings that preserve the word context, i.e., Word2Vec, FastText, and GloVe, pre-trained and trained on our dataset to vectorize the preprocessed dataset. For the misinformation task, we train a Logistic Regression as a baseline and compare its results with the performance of ten Deep Learning architectures. We obtain the best results using a Recurrent Convolutional Neural Network based architecture. The experimental results show that the models are highly dependable on text preprocessing and the word embedding employed.

No abstract

Extracting top-k keywords and documents using weighting schemes are popular techniques employed in text mining and machine learning for different analysis and retrieval tasks. The weights are usually computed in the data preprocessing step, as they are costly to update and keep track of all the modifications performed on the dataset. Furthermore, computation errors are introduced when analyzing only subsets of the dataset. Therefore, in a Big Data context, it is crucial to lower the runtime of computing weighting schemes, without hindering the analysis process and the accuracy of the machine learning algorithms. To address this requirement for the task of top-k keywords and documents, it is customary to design benchmarks that compare weighting schemes within various configurations of distributed frameworks and database management systems. Thus, we propose a generic document-oriented benchmark for storing textual data and constructing weighting schemes (TextBenDS). Our benchmark offers a generic data model designed with a multidimensional approach for storing text documents. We also propose using aggregation queries with various complexities and selectivities for constructing term weighting schemes, that are utilized in extracting top-k keywords and documents. We evaluate the computing performance of the queries on several distributed environments set within the Apache Hadoop ecosystem. Our experimental results provide interesting insights. As an example, MongoDB proves to have the best overall performance, while Spark's execution time remains almost the same, regardless of the weighting schemes.

No abstract

Cloud Systems provide the computing infrastructure and on-demand capacity required to host services.In this paper we present a new provisioning mechanism for Cloud Systems. Our project addresses the key requirements in managing resources at the infrastructure level. The proposed resource manager allocates virtual resources in a flexible way, taking into account time, costs and physical resources. It provides on-demand elasticity, which is one of the most important feature of Cloud Computing towards the traditional hosting strategies. New resources can be dynamically allocated ondemand or policy-based. This is one of the novel contribution of this paper, as the majority of the cloud resource managers provides just static allocation.For resource scheduling we have developed a mechanism that takes into account virtual machines' capabilities and well defined policies and it uses a genetic algorithm. We used a QoS constraint base algorithm in order to maximize the performance, according to different defined policies.In order to demonstrate the performance of the presented resource managing mechanism we provide and interpret several experimental results.