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.

Top-k high utility pattern mining, which extracts the highest top-k patterns that the users want to find, has been actively studied. Most previous studies in this domain have focused on static databases, where data insertions do not occur. In the real world, however, various applications continuously generate new data, and existing top-k high utility pattern mining algorithms devised to process static databases cannot handle incremental databases. Although some methods can handle stream data, they have the limitation of processing a portion of the database rather than the entire accumulated database. In this paper, we suggest an efficient incremental mining method that discovers top-k high utility patterns from the entire accumulated database. The proposed approach utilizes a list structure that stores minimal utility information required for the mining process and does not generate candidate itemsets. The suggested algorithm processes the incremental data with a single database scan and restructures the list for efficient mining. Moreover, four efficient threshold raising techniques along with a restoring technique are utilized to calculate the optimal threshold value in an accumulated incremental environment. The results of the experiments on runtime, memory, and scalability show that the suggested method efficiently processes the entire incremental database.

This paper presents a dataset comprising 700 video sequences encoded in the two most popular video formats (codecs) of today, H.264 and H.265 (HEVC). Six reference sequences were encoded under different quality profiles, including several bitrates and resolutions, and were affected by various packet loss rates. Subsequently, the image quality of encoded video sequences was assessed by subjective, as well as objective, evaluation. Therefore, the enclosed spreadsheet contains results of both assessment approaches in a form of MOS (Mean Opinion Score) delivered by the absolute category ranking (ACR) procedure, SSIM (Structural Similarity Index Measure) and VMAF (Video Multimethod Assessment Fusion). All assessments are available for each test sequence. This allows a comprehensive evaluation of coding efficiency under different test scenarios without the necessity of real observers or a secure laboratory environment, as recommended by the ITU (International Telecommunication Union). As there is currently no standardized mapping function between the results of subjective and objective methods, this dataset can also be used to design and verify experimental machine learning algorithms that contribute to solving the relevant research issues.

Long-term urban traffic flow prediction is an important task in the field of intelligent transportation, as it can help optimize traffic management and improve travel efficiency. To improve prediction accuracy, a crucial issue is how to model spatiotemporal dependency in urban traffic data. In recent years, many studies have adopted spatiotemporal neural networks to extract key information from traffic data. However, most models ignore the semantic spatial similarity between long-distance areas when mining spatial dependency. They also ignore the impact of predicted time steps on the next unpredicted time step for making long-term predictions. Moreover, these models lack a comprehensive data embedding process to represent complex spatiotemporal dependency. This paper proposes a multi-scale persistent spatiotemporal transformer (MSPSTT) model to perform accurate long-term traffic flow prediction in cities. MSPSTT adopts an encoder-decoder structure and incorporates temporal, periodic, and spatial features to fully embed urban traffic data to address these issues. The model consists of a spatiotemporal encoder and a spatiotemporal decoder, which rely on temporal, geospatial, and semantic space multi-head attention modules to dynamically extract temporal, geospatial, and semantic characteristics. The spatiotemporal decoder combines the context information provided by the encoder, integrates the predicted time step information, and is iteratively updated to learn the correlation between different time steps in the broader time range to improve the model's accuracy for long-term prediction. Experiments on four public transportation datasets demonstrate that MSPSTT outperforms the existing models by up to 9.5% on three common metrics.

IntroductionPsychoacoustic measures are used to quantify tinnitus perception in clinical and research settings. Residual inhibition (RI), for instance, is defined as a temporary reduction in tinnitus loudness after the presentation of an acoustic stimulus. A novel approach was developed recently to assess the minimum sound level producing RI, the minimum residual inhibition level (MRIL), after the repeated presentation of short sounds separated by short periods of silence. Two studies were performed: the objective of the study 1 was to assess the reliability and the reproducibility of this measures as well as others, including the minimum masking level (MML). The objective of study 2 was to assess whether a smoother envelope of the sound would produce similar levels of MML and MRIL and improve listening comfort.MethodsThirty participants with chronic tinnitus completed each study. For study 1, MML and MRIL were obtained at the seven standard audiometric frequencies at around 1-month interval. For study 2, MML and MRIL were obtained using triangular and trapezoidal-shaped NBN signals centered at three different frequencies.ResultsThe mean absolute test-retest difference was 3 dB for both MML and MRIL [Intraclass correlation, ICC (2,1) of 0.70 and 0.84, respectively] for the measures obtained in test and retest (86% of cases for MML and 21% of cases for complete MRIL). While the two types of envelopes provided comparable MRIL, the “triangular” envelope provided more listening comfort.DiscussionIn sum, the novel approach provides quick and reliable MML and MRIL measurement, with the potential benefit of subtyping participants.