Search

Disinformation, often known as fake news, is a major issue that has received a lot of attention lately. Many researchers have proposed effective means of detecting and addressing it. Current machine and deep learning based methodologies for classification/detection of fake news are content-based, network (propagation) based, or multimodal methods that combine both textual and visual information. We introduce here a framework, called FNACSPM, based on sequential pattern mining (SPM), for fake news analysis and classification. In this framework, six publicly available datasets, containing a diverse range of fake and real news, and their combination, are first transformed into a proper format. Then, algorithms for SPM are applied to the transformed datasets to extract frequent patterns (and rules) of words, phrases, or linguistic features. The obtained patterns capture distinctive characteristics associated with fake or real news content, providing valuable insights into the underlying structures and commonalities of misinformation. Subsequently, the discovered frequent patterns are used as features for fake news classification. This framework is evaluated with eight classifiers, and their performance is assessed with various metrics. Extensive experiments were performed and obtained results show that FNACSPM outperformed other state-of-the-art approaches for fake news classification, and that it expedites the classification task with high accuracy.

The laser welding technique is quite common in various manufacturing lines, including those for lithium-ion power batteries, due to its remarkable productivity, effectiveness, and flexibility. However, it has been observed that the consistency of the welding quality is not always optimal. This study investigates the adoption of laser welding in the manufacturing of lithium-ion batteries, such as their anode, cathode and safety vent, where consistent weld quality is crucial for battery performance and safety. Therefore, evaluating the laser-welding product is indispensable for industrial production and the public domain. Several techniques have been utilized for laser welding evaluation, both destructive and non-destructive. However, these methods are ineffective and too cumbersome to be adopted in mass manufacturing. In contrast, a machine vision strategy has recently been adopted to distinguish between successful and unsuccessful laser welding items. This opened new perspectives for evaluating the quality of the laser welding product using digital image techniques. However, these methods cannot deliver outstanding performance across multiple laser-welding products and achieve optimal classification accuracy. Deep learning has evolved remarkably in recent years and gained popularity in detecting welding defects. This paper presents the observation of a Hybrid Vision Transformer (HViT) to classify the laser welding images corresponding to their feature patch images based on multi-model feature aggregation. The VGG-16 and MobileNet, which were pre-trained on ImageNet, were utilized as the core models to extract the rich features of the laser welding image. To integrate these features, the squeeze excitation module (SE) was utilized, and a multi-layer perception (MLP) approach with a label smoothing optimizer was used for classification. To determine the effectiveness of the proposed strategy, a comparative analysis was conducted with a multitude of alternative machine learning and deep learning approaches. The results indicate that the proposed strategy surpasses all other methods in terms of classification accuracy and evaluation metrics on the test dataset.

The Spike (S) protein of the SARS-CoV-2 virus that causes the COVID-19 disease is considered the most important target for vaccine, drug and therapeutic research as it attaches and binds to the ACE2 receptor of the host cells and allows the entry of this virus. Analysis and classification of newly determined S protein structures for SARS-CoV-2 are critical to properly understand their functional, evolutionary and architectural relatedness to already known protein structures. In this paper, first, the comparative analysis of SARS-CoV-2 S protein structures is performed. Through comparative analysis, the S protein structures in the PDB (protein data bank) database are compared and analyzed not only with each other but with the structures of other viruses for various parameters. Second, the S protein structures in PDB are classified into different variants, and the associated published literature is studied to investigate what kind of therapeutics (antibodies, T-cell receptors and small molecules) are used on the structures. This is the first study that classifies the S protein structures of the SARS-CoV-2 in PDB into various variants, and the obtained comparative analysis results could be beneficial to the research community, in general, and to crystallographers and health workers, in particular.

Discovering frequent itemsets is a data analysis task used in numerous domains. It consists of finding sets of items (itemsets) that frequently appear in a set of database records (also called transactions). Though discovering frequent itemsets is useful, it can produce a large amount of spurious patterns. As a result, the user may spend a great amount of time to analyze the itemsets found by a frequent itemset mining algorithm to find truly interesting patterns. Hence, in recent years, a key research topic has emerged which is to discover statistically significant patterns in databases. The most popular model for identifying itemsets that are statistically significant is to discover non-redundant productive itemsets. The state-of-the-art algorithm to extract this set of patterns is OPUS-Miner. A key drawback of that algorithm is that it is designed to be applied to a static database. Moreover, a second drawback of OPUS-Miner is that it discovers all patterns in a database. In other words, the user cannot search for itemsets containing some specific items. This paper addresses these issues by defining the novel problem of discovering targeted non redundant productive itemsets in dynamic databases. An algorithm named IDPI+ (Interactive Discovery of Productive Itemsets) is presented, storing transactions in a tree structure, which can then be interactively queried to identify productive and non redundant itemsets containing specific items. A structure named Query-Tree is also introduced to process many queries at the same time. Moreover, to handle dynamic databases, efficient transaction insertion and deletion algorithms are provided to update the tree. It was observed in an experimental evaluation on benchmark datasets containing various types of data that IDPI+ can handle thousands of queries per second on a desktop computer. Moreover, it was found that IPDI+ is more than an order of magnitude faster than a baseline algorithm.

Interactive theorem provers (ITPs) are software tools that allow human users to write and verify formal proofs. In recent years, an emerging research area in ITPs is proof mining, which consists of identifying interesting proof patterns that can be used to guide the interactive proof process in ITPs. In previous studies, some data mining techniques, such as frequent pattern mining, have been used to analyze proofs to find frequent proof steps. Though useful, such models ignore the facts that not all proof steps are equally important. To address this issue, this paper proposes a novel proof mining approach based on finding not only frequent patterns but also high utility patterns to find proof steps of high importance (utility). A proof process learning approach is proposed based on high utility itemset mining (HUIM) for the PVS (Prototype Verification System) proof assistant. Proofs in PVS theories are first abstracted to a computer-processable corpus, where each line represents a proof sequence and proof commands in proof sequences are associated with utilities representing their weightage (importance). HUIM techniques are then applied on the corpus to discover frequent proof steps/high utility patterns and their relationships with each other. Experimental results suggest that combining frequent pattern mining techniques, such as sequential pattern mining and high utility itemset mining, with proof assistants, such as PVS, is useful to learn and guide the proof development process.

Generally, two topics with vastly different terminology probably indicate different implied concepts. However, these topics themselves might share common references (bibliographic coupling), which suggest the underlying joint concept. Therefore, searching for these joint concepts in different topics would be of scientific interest. Previous studies have measured the similarity between topics based on comparison of the topics' word probability distributions. In contrast, this paper presents an approach for measuring the similarity between topics based on the bibliographic coupling. Besides, the similarity is independent of the topic's word probability distributions generated by a Latent Dirichlet Allocation (LDA) model. The proposed approach was evaluated using its counterpart (intra-topic similarity), baseline topic similarity matrices, and cosine measure. The method was exampled on brain cancer patents. A cross-topic similarity network of eight topics showcases 28 cross-topic pairs to profile which topics were associated with particular topics. Interestingly, some of the 28 combinations may be of scientific interest. For instance, the findings of the top five cross-topic pairs suggest that “growth of cancer cells” and “imbalances in the hormones” have common knowledge sources with the highest similarity value. These two entirely different concepts may suggest some common causative factors within the field. We believe that finding such an association between unrelated innovative inventions across various industries may help public and private research units in planning research direction and serve as a reference for future research.

Social online marketing is expanding fast with the evolution and recent development in the information and communication technology. Investigating how companies are exploiting social media for marketing, advertisement, and consumer's engagement is gaining more and more interest. In this paper, brands/companies data on Twitter is collected and analyzed to compute the overall company response on Twitter. Responsiveness of a company is inferred from three features: company popularity, average company replies, and average followers' replies. Twitter network features are used in calculating the posting frequency for companies and their followers. It is shown that the proposed approach can be used in finding the responsiveness of companies and their followers. Furthermore, useful links for a brand consumer is extracted and the posting behavior of brands and their followers is determined with the help of Twitter network features, such as retweet count and geolocation. This paper contributes to the literature on how Twitter data and its network structure features can be exploited in finding the responsiveness and posting behavior of companies and their followers. We believe that this approach can be used effectively in developing prediction and information-filtering systems, particularly the personalized-recommendation systems.