A Microblog Popularity Prediction Model Based on Temporal Sequence Features and Text Features

Rumor detection is a critical task for addressing the spread of misinformation and maintaining the credibility of information sources. Natural Language Processing (NLP) techniques have been employed to propose efficient and effective methods for rumor detection. In the wake of the widespread COVID-19 pandemic, the world has faced extensive strain on health, economics, and social structures. The dissemination of false or inaccurate information on social media, whether intentionally malicious or unintentional, has had detrimental consequences for individuals and society, particularly during critical situations like real world emergencies. In this study, we aim to explore the textual and temporal features present in social media posts (specifically tweets) related to COVID-19 to detect rumors as time is unique feature of text and any event can be mapped on timeline. Previous studies utilized the textual features and the temporal features are neglected at large for rumors detection. We utilize both temporal and textual features independently, as well as in combination, to train machine learning and neural network models. The evaluation of multiple algorithms (RNN, LSTM, CNN, DNN, BERT) across various feature sets reveals diverse performance. RNN and LSTM improve with combined textual and temporal features, highlighting temporal information’s importance. CNN performs well with textual features but declines with temporal features. DNN maintains consistent performance, while BERT demonstrates moderate effectiveness in classification tasks.

Topic popularity prediction is an important task on social media, which aims at predicting the ongoing trends of topics according to logged historical text-based records. However, only limited existing approaches apply sentiment analysis to facilitate popularity prediction. Public sentiment is worth taking into consideration because the topics with strong sentiment tend to spread faster and broader on social media. In this paper, we propose a novel framework, SENTI2POP, to predict topic popularity utilizing sentiment information. We first adapt a state-of-art popularity quantification method to capture the topic popularity, and then design a novel tree-like network (Tree-Net) combining Long Short-Term Memory (LSTM) and Convolutional Neural Network (CNN) for sentiment analysis. In addition, we propose a sentiment-aware time series prediction approach based on Dynamic Time Warping (DTW) and Autoregressive Integrated Moving Average model (ARIMA) to predict topic popularity. We prove by experiments that SENTI2POP outperforms the existing popularity prediction models on a real-world Twitter dataset by reducing the prediction error. Experimental results also show that SENTI2POP could be applied to improve the accuracy of most non-sentiment popularity prediction models.

Analyzing the dynamics of information diffusion cascades and accurately predicting their behavior holds significant importance in various applications. In this paper, we concentrate specifically on a recently introduced contrastive cascade graph learning framework, for the task of predicting cascade popularity. This framework follows a pre-training and fine-tuning paradigm to address cascade prediction tasks. In a previous study, the transferability of pre-trained models within the contrastive cascade graph learning framework was examined solely between two social media datasets. However, in our present study, we comprehensively evaluate the transferability of pre-trained models across 13 real datasets and six synthetic datasets. We construct several pre-trained models using real cascades and synthetic cascades generated by the independent cascade model and the Profile model. Then, we fine-tune these pre-trained models on real cascade datasets and evaluate their prediction accuracy based on the mean squared logarithmic error. The main findings derived from our results are as follows. (1) The pre-trained models exhibit transferability across diverse types of real datasets in different domains, encompassing different languages, social media platforms, and diffusion time scales. (2) Synthetic cascade data prove effective for pre-training purposes. The pre-trained models constructed with synthetic cascade data demonstrate comparable effectiveness to those constructed using real data. (3) Synthetic cascade data prove beneficial for fine-tuning the contrastive cascade graph learning models and training other state-of-the-art popularity prediction models. Models trained using a combination of real and synthetic cascades yield significantly lower mean squared logarithmic error compared to those trained solely on real cascades. Our findings affirm the effectiveness of synthetic cascade data in enhancing the accuracy of cascade popularity prediction.

Social media popularity prediction is an important channel to explore content sharing and communication on social networks. It aims to capture informative cues by analyzing multi-type data (such as user profile, image, and text) to decide the popularity of a specified post. In this article, we divide social network users into two categories (i.e., active and inactive users) and find a dilemma in existing models: If an active user publishes the low-popularity post, the model will habitually predict the high score. On the contrary, if an inactive user provides the high-popularity post, the model still gives the low score incorrectly. Therefore, how to make the model more subtle to users is important. Comparing to existing methods that directly leverage multi-modal features for regression training, this article stresses more on two novel mechanisms. The first method aims to prevent the over-fitting on user IDs. We propose the attribute-sensitive interactive mechanism (M1) by incorporating explicit user-attribute and post-attribute interaction. It can analyze which type of features a user cares the most and weaken the model’s dependence on user IDs. The second method aims to strengthen the influence of post content. We propose the knowledge embedding mechanism (M2) to revise the popularity scores in existing models by fusing the statistical frequency over multi-type data. Note that both mechanisms are model-agnostic, which can be applicable in any popularity prediction model. Extensive experiments conducted on the Social Media Prediction Dataset further validate the effectiveness.

Depression screening with textual and audio features based on large language models and machine learning.

Detecting fake news and missing information is gaining popularity, especially after social media and online news platforms advancements. Social media is the main and speediest source of fake news propagation, whereas online news websites contribute to fake news dissipation. In this study, we propose a framework to detect fake news using the temporal features of text and consider user feedback to determine whether the news is fake or not. In recent studies, the temporal features in text documents gain valuable consideration from Natural Language Processing and user feedback and only try to classify the textual data as fake or true. This research article indicates the impact of recurring and non-recurring events on fake and true news. We use different models such as LSTM, BERT, and CNN- BiLSTM to investigate, and it is concluded that from BERT, we get better results, and 70% of true news is recurring, and the rest of 30% is non-recurring.

Timely and accurate traffic flow information is essential for the realization of intelligent transportation systems (ITS). For the existing traffic flow prediction models, only the temporal features were extracted but the spatial features were neglects. A short-term traffic flow prediction model based on Pearson Correlation Coefficient (PCC) and Bi-directional Long Short-Term Memory Network (BiLSTM) model was proposed. The PCC was used to mine the spatial dependence of the road network, and the temporal features of traffic flow were mined using BiLSTM model. The extracted temporal and spatial features were combined to achieve short-term flow prediction. The experimental results show that the PCC-BiLSTM model has higher prediction accuracy and lower error than the prediction models considering only temporal features, which verifies the effectiveness of traffic flow prediction in consideration of both temporal and spatial characteristics.

In social networks, users can easily share information and express their opinions. Given the huge amount of data posted by many users, it is difficult to search for relevant information. In addition to individual posts, it would be useful if we can recommend groups of people with similar interests. Past studies on user preference learning focused on single-modal features such as review contents or demographic information of users. However, such information is usually not easy to obtain in most social media without explicit user feedback. In this paper, we propose a multimodal feature fusion approach to implicit user preference prediction which combines text and image features from user posts for recommending similar users in social media. First, we use the convolutional neural network (CNN) and TextCNN models to extract image and text features, respectively. Then, these features are combined using early and late fusion methods as a representation of user preferences. Lastly, a list of users with the most similar preferences are recommended. The experimental results on real-world Instagram data show that the best performance can be achieved when we apply late fusion of individual classification results for images and texts, with the best average top-k accuracy of 0.491. This validates the effectiveness of utilizing deep learning methods for fusing multimodal features to represent social user preferences. Further investigation is needed to verify the performance in different types of social media.

With the development of online healthcare social media, a large volume of user generated content are available in different health discussion forums or groups. Healthcare social media sites could empower patients to play a substantial role in their treatment by acquiring knowledge and support through actively involving in online discussions and interactions. However, users may have difficulty to find relevant topics or peers in these online health forums with a large amount of unstructured information. Most recommendation systems rely on content-based approach to recommend peers or discussions to their participants. However, in healthcare social media sites, content-based approach is not sufficient because health consumers may have different intentions of participation or may be interest in different types of support even if the content matches their interest. Based on previous studies, we utilize Naïve Bayes methods and propose two tasks for classifying posts and comments on Quit Stop forum, an online community for smoking cessation intervention, respectively: (1) classification of intentions and (2) classification of social support types. Different text feature sets and user health feature sets are selected to develop classifiers. Taking different evaluation indicators as optimizing goals, we develop genetic algorithms to combine classifiers with different feature sets and optimize the classification results. It is found that for post classification, integrating text and health features could achieve the highest precision, recall and F1 measure. For comment classification, combining different text features could reach the best result. In the future, the classification result could be applied to developing recommender systems for topic recommendation and user prediction of online health forums.

Background Major component of ACS, Non-ST segment elevation myocardial infarction (NSTEMI), has more than twice the incidence compared to ST-segment elevation myocardial infarction (STEMI). GRACE (Global Registry of Acute Coronary Events) and TIMI (Thrombolysis in Myocardial Infarction) risk scores are the two most popular prediction models in ACS due to TIMI’s advantage in its simplicity in clinical application and GRACE’s favorable discriminating power. Nevertheless, limited evidence compared the usefulness of these two major risk scores and their association with in-hospital events. Purpose To compare between the predictive ability of GRACE & TIMI risk scores for In hospital outcomes among the NSTEMI patients . Methods This prospective observational study was carried out from 8th march 2020 to 7th March 2022. Total 2152 acute NSTEMI patients were included. Both GRACE & TIMI risk score were calculated. For GRACE score, patients who scored ≤108 points were low risk,109-139 points were medium risk, ≥140 points were high risk . For TIMI scores patients with 0-2 scores were low risk , 3-4 scores were medium risk and 5-7 score were High risk. In Hospital outcomes were defined as the in hospital composite events which includes acute left ventricular failure, cardiogenic shock, arrhythmia & mortality. Results Among the 2152 patients majority were male (73.5%) with mean age of the study population were 58.33 ± .25 years. Mean Grace score & TIMI score of the study population were 137.81±16.38& 3.29±.97 respectively. Most of the patients were categorized in to Medium risk groups by both GRACE & TIMI risk scores , 70% & 51.3% respectively. However, high risk group percentage were higher in GRACE score group(41.3%) in comparison to TIMI score group(10.3%). GRACE score showed better predictive ability than TIMI score for in-hospital outcomes as evaluated by ROC curve( AUC of GRACE vs TIMI, In hospital ALVF:0.89vs 0.59, P<0.001; Cardiogenic Shock 0.72 vs 0.57, P=0.002; Mortality: 0.75 vs 0.59, P=0.002; arrhythmia 0.72 vs0.63, P=0.003). Conclusion GRACE showed better predictive accuracy than TIMI in NSTEMI patients for in-hospital outcomes. The sequential use of TIMI and GRACE scores provide an easy and promising discriminative tool in predicting outcomes in NSTEMI patients , thus improve the identification of high-risk patients.Pic :ROC curve for prediction of ALVFPic: ROC curve for Mortality prediction

This work investigates the impact of patients' heterogeneity, including factors such as gender, age, and clinical conditions, on the performance of machine learning models in predicting blood glucose levels in individuals with Type 1 Diabetes. Using the T1DiabetesGranada dataset, various datasets were generated based on these heterogeneity factors. Three popular prediction models -a Linear model, an LSTM neural network, and a CNN- were applied to both the generated datasets and the complete dataset to measure prediction performance at a 30-minute prediction horizon. The preliminary results suggest that incorporating patient-specific heterogeneity factors generally improves prediction performance, highlighting the existence of bias in standard blood glucose level prediction models. Future research should explore whether these findings hold in other related datasets.

With the rapid advancements in intelligent transportation systems Jiang et al. (In: In Proceedings of the 30th ACM international conference on information & knowledge management 4515-4525, 2021), (Feng et al. Digit Commun Networks, 2024), precisely forecasting traffic information has emerged as a significant challenge. Recently, numerous advanced neural networks with complex architectures have been introduced to address this challenge. Nonetheless, the majority of these models handle temporal and spatial features separately before combining them, which overlooks the inherent relationships between these two types of characteristics. This approach of independent feature extraction can result in the loss of valuable information and restricts the model's capacity to effectively leverage the interdependencies between spatial and temporal features. In order to tackle this challenge, we introduce STIC, a Transformer-based neural network designed to capture crucial information from both spatial and temporal domains. The main innovation of our method lies in utilizing the cross-attention mechanism within Transformers to sequentially capture and adaptively merge spatiotemporal features from historical data. Experiments conducted on four diverse traffic forecasting datasets show that our model outperforms traditional methods by effectively uncovering the underlying spatial and temporal dependencies in traffic data sequences. Our work introduces a new strategy for enhancing the accuracy of traffic flow predictions.

Social media, in recent years, have become an invaluable source of information for both public and private organizations to enhance the comprehension of people interests and the onset of new events. Twitter, especially, allows a fast spread of news and events happening real time that can contribute to situation awareness during emergency situations, but also to understand trending topics of a period. The article proposes an online algorithm that incrementally groups tweet streams into clusters. The approach summarizes the examined tweets into the cluster centroid by maintaining a number of textual and temporal features that allow the method to effectively discover groups of interest on particular themes. Experiments on messages posted by users addressing different issues, and a comparison with state-of-the-art approaches show that the method is capable to detect discussions regarding topics of interest, but also to distinguish bursty events revealed by a sudden spreading of attention on messages published by users.

3 - Time varying signals

Multiple modalities represent different aspects by which information is conveyed by a data source. Modern day social media platforms are one of the primary sources of multimodal data, where users use different modes of expression by posting textual as well as multimedia content such as images and videos for sharing information. Multimodal information embedded in such posts could be useful in predicting their popularity. To the best of our knowledge, no such multimodal dataset exists for the prediction of social media photos. In this work, we propose a multimodal dataset consisiting of content, context, and social information for popularity prediction. Speci?cally, we augment the SMPT1 dataset for social media prediction in ACM Multimedia grand challenge 2017 with image content, titles, descriptions, and tags. Next, in this paper, we propose a multimodal approach which exploits visual features (i.e., content information), textual features (i.e., contextual information), and social features (e.g., average views and group counts) to predict popularity of social media photos in terms of view counts. Experimental results con?rm that despite our multimodal approach uses the half of the training dataset from SMP-T1, it achieves comparable performance with that of state-of-the-art.