Search

Accurately predicting the future trajectory of pedestrians is critical for tasks such as autonomous driving and robot navigation. Previous methods for pedestrian trajectory prediction dealt with social interaction and pedestrian movement factors either concurrently or sequentially, neglecting the link between them. Therefore, a spatial–temporal interaction-aware graph convolution network (STIGCN) is proposed for pedestrian trajectory prediction. STIGCN considers the correlation between social interaction and pedestrian movement factors to achieve more accurate interaction modeling. Specifically, we first constructed spatial and temporal graphs to model social interactions and movement factors. Then, we designed the spatial–temporal interaction-aware learning to utilize the spatial interaction features of each moment to assist the temporal interaction modeling and utilize the temporal interaction features of each pedestrian to assist the spatial interaction modeling, resulting in more accurate interaction modeling. Finally, a time-extrapolator pyramid convolution neural network (TEP-CNN) is designed to jointly estimate the two-dimensional Gaussian distribution parameters of future trajectories by combining the prediction features from multiple layers. Experimental results on two benchmark pedestrian trajectory prediction datasets show that our proposed method outperforms existing methods in terms of average displacement error and final displacement error and achieves more accurate predictions for pedestrian motions such as convergence and encounter. [ABSTRACT FROM AUTHOR]

The increasing popularity of social media platforms makes it important to study user engagement, which is a crucial aspect of any marketing strategy or business model. The oversaturation of image content on social media platforms has persuaded us to identify the important factors that affect the popularity of image content in social media. This comes from the fact that only an iota of the humongous image content available online receives the attention of the target audience. A good quality image having the potential content to make the image popular, can be used in online business, to attract potential buyers. Comprehensive research has been done in the area of popularity prediction using several machine learning techniques. However, we observe that there is still significant scope for improvement in analyzing the social importance of image content. We propose the DFW-PP framework, to learn the importance of different features obtained from the image content, that vary over time. The proposed method extracts features from the image content from social media, provides scores to the image content based on studies of the dynamics of the image content, and draws inferences on the impact of the image on social media. Further, the proposed method controls the skewness of the distribution of the features by applying a log–log normalization. The proposed method is experimented with a benchmark dataset, to show promising results. Our statistical analysis shows a significant influence of a combination of image and meta features on the popularity of the image in social media. The code will be made publicly available at github. [ABSTRACT FROM AUTHOR]

Understanding the dynamic nature of group formation and evolution in social networks is seen as a significant step to better describe how individuals' trust nodes are gathered and form communities. There has been considerable growth in interest in the modeling and evaluation of signed social networks (SSNs) and extracting their essential properties to improve their trustiness. In order to improve the trustiness of these networks, we propose a differential machine learning-based approach to predict trust values in SSNs, called diffTrbML. Differential machine learning considers inputs, labels, and differentials of labels to inputs to enhance training performance. By training on collected real-world datasets customized by trust feature vectors for SSNs, a trust value predictor is established to calculate the node trust value. We also evaluate prediction results using common machine learning measures including precision, recall, F1 score, accuracy, and RMSE on two real-world SSNs for nodes based on trust values. Lastly, diffTrbML resulted in significantly greater levels of learning measures when customized features for SSNs were investigated. The experiments in large social networks show that the proposed trust prediction method can provide a high-performance learning approach. [ABSTRACT FROM AUTHOR]

Sarcasm is a type of speech where people use positive words to convey a negative message. Recently, only a few research have been presented that focus on the entire spectrum of sarcasm in order to identify sarcastic sentiments present in both the image and the text. This work presents a novel firebug swarm optimization-based long short-term memory (FSO-LSTM) architecture to identify the sarcastic sentiments present in tweets. To identify the facial expressions of the users, the proposed FSO-based LSTM architecture is trained using the CK + dataset. The FSO algorithm is used to optimize the weighting factors of the LSTM architecture and also minimize the root-mean-square error (RMSE) and mean absolute error. The proposed method primarily attempts to address two challenging issues in sarcasm detection: the high number of false negatives and the fact that polite tweets often go undetected. The user's mood changes (sarcastic) such as rude, polite, furious, and impassive can be identified using the proposed model. Hence, the proposed classifier is capable of analyzing the behavior change of the user by collecting the past twitter account history. The efficiency of the proposed methodology is evaluated using different performance metrics such as accuracy, RMSE, confusion matrix, and loss. The proposed methodology offers an average classification accuracy of 97.25% classification accuracy when compared to the state-of-the-art approaches. [ABSTRACT FROM AUTHOR]

Today, social networks have created a wide variety of relationships between users. Friendships on Facebook and trust in the Epinions network are examples of these relationships. Most social media research has often focused on positive interpersonal relationships, such as friendships. However, in many real-world applications, there are also networks of negative relationships whose communication between users is either distrustful or hostile in nature. Such networks are called signed networks. In this work, sign prediction is made based on existing links between nodes. However, in real signed networks, links between nodes are usually sparse and sometimes absent. Therefore, existing methods are not appropriate to address the challenges of accurate sign prediction. To address the sparsity problem, this work aims to propose a method to predict the sign of positive and negative links based on clustering and collaborative filtering methods. Network clustering is done in such a way that the number of negative links between the clusters and the number of positive links within the clusters are as large as possible. As a result, the clusters are as close as possible to social balance. The main contribution of this work is using clustering and collaborative filtering methods, as well as proposing a new similarity criterion, to overcome the data sparseness problem and predict the unknown sign of links. Evaluations on the Epinions network have shown that the prediction accuracy of the proposed method has improved by 8% compared to previous studies. [ABSTRACT FROM AUTHOR]

Link prediction (LP) is a crucial issue in the online social network (OSN) evolution analysis. Since OSNs are growing in size on a daily basis, a growing need for scalable LP algorithms is being felt. OSNs are innately evolutionary, such that the characteristics, behavior, and activities of their components (including nodes and links) change over time. In analyzing social networks which are based on the time evolution model, LP helps us realize the logic of social network growth. Deriving time patterns of evolutionary changes according to the communities and neighbors of nodes in a network can be aptly used for LP. This article introduces a new algorithm based on irregular cellular learning automata (ICLAs) for LP in the near future in OSNs. The algorithm we propose here models the network as an ICLA. The ICLA weighs the real links in the network according to entities' participation in forming communities over consecutive time periods. This method lies in the premise that social networks include communities. Based on the communities formed over successive time periods, the presented method calculates the probability of link formation between every pair of nodes which are unconnected at the present time, estimating the chances of their connection in the near future. Experiments performed on real social networks show that the proposed algorithm produces good results in predicting link formation in OSNs. [ABSTRACT FROM AUTHOR]

User-item rating interactions in the recommender system have a deep potential connection with the friend relationships in the social network. In short, users who like the same kind of items may be potential friends in social network, and vice versa, friends in social networks tend to like similar items. Although the above-mentioned two kinds of interactive information can complement and inspire each other, either of them is sparse, which is still not enough to make accurate recommendations. In order to make up for this defect, we then mine useful information from attribute information, learning more informative node representation. In this paper, we explore attribute learning and mutual utilization, complementation and inspiration between social data and rating data. We propose a generic Attribute-Aware Multi-task Recommendation framework (AAMR) for rating prediction and social prediction, which learns representations for users and items by preserving both rating data and social data and attribute information, so as to conduct both rating prediction and trust relationship prediction tasks. Because many users are both in the rating matrix and in social networks, in the common learning, the two tasks will share the embedding of users, which makes the social data and rating data enrich each other's semantics and alleviate each other's sparsity. To justify our proposal, we conduct extensive experiments on a real-world dataset. Compared to the state-of-the-art rating and trust prediction approaches, AAMR can learn more informative representations, achieving substantial gains on both tasks. [ABSTRACT FROM AUTHOR]