We will review some of the major results in random graphs and some of the more challenging open problems. We will cover algorithmic and structural questions. We will touch on newer models, including those related to the WWW.

Random graphs

2011 international joint conference on neural networks (ijcnn)

IoT-based WSNs have proved their significance in delivering critical information pertaining to hostile applications such as Wildfire Detection (WD) with the least possible delay. However, the sensor nodes deployed in such networks suffer from the perturbing concern of limited energy resources, restricting their potential in the successful detection of wildfire. To extenuate this concern, we propose an intelligent framework, Sleep scheduling-based Energy Optimized Framework (SEOF), that works in two folds. Firstly, we propose an energy-efficient Cluster Head (CH) selection employing a recently developed meta-heuristic method, Tunicate Swarm Algorithm (TSA), that optimizes the five novel fitness parameters by integrating them into its weighted fitness function. Secondly, we perform a sleep scheduling of closely-located sensor nodes based on the distance threshold calculated through a set of experiments. Sleep scheduling methodology plays a pivotal role in abating the number of data transmissions in SEOF. Finally, we simulate SEOF in MATLAB under different scenarios to examine its efficacy for the various performance metrics and scalability features. Our empirical results prove that SEOF has ameliorated the network stability period for two different scenarios of network parameters by 35.3% and 216% vis-a-vis CIRP.

/pdf/intelligent-framework-using-iot-based-wsns-for-wildfire-nuw982lfcq.pdf

Intelligent Framework Using IoT-Based WSNs for Wildfire Detection

Users in online networks exert different influence during the process of information propagation, and the heterogeneous influence may contribute to personalized recommendations. In this paper, we analyse the topology of social networks to investigate users’ influence strength on their neighbours. We also exploit the user-item rating matrix to find the importance of users’ ratings and determine their influence on entire social networks. Based on the local influence between users and global influence over the whole network, we propose a recommendation method with indirect interactions that makes adequate use of users’ relationships on social networks and users’ rating data. The two kinds of influence are incorporated into a matrix factorization framework. We also consider indirect interactions between users who do not have direct links with each other. Experimental results on two real-world datasets demonstrate that our proposed framework performs better than other state-of-the-art methods for all users and cold-start users. Compared with node degrees, betweenness, and clustering coefficients, coreness constitutes the best topological descriptor to identify users’ local influence, and recommendations with the measure of coreness outperform other descriptors of user influence.

/pdf/topological-influence-aware-recommendation-on-social-3kp5awl7bq.pdf

Topological Influence-Aware Recommendation on Social Networks

The technological progression in the area of Wireless Body Area Network (WBAN) has made it possible to design various sensing modules operating for detecting the multitudinous physical attributes of a patient's body. However, the limited battery of these sensing devices has restricted the scope for WBAN. Hence, it is imperative to design an energy efficient algorithm to combat the gigantic energy consumption of these sensor nodes. Therefore, in this paper, we propose a Novel Energy Efficient hybrid Meta‐Heuristic Approach (NEEMA) for WBAN. We adopt this hybrid approach by using Tunicate Swarm Algorithm (TSA) and Genetic Algorithm (GA); we name it as T‐GA, to deliver the high convergence and large exploitation and exploration capabilities. We follow the clustering approach by selecting the Cluster Head (CH) with the help of the fitness function of T‐GA. We consider the novel parameters for selecting the CH that helps in the overall energy preservation of sensor nodes. Our proposed work focuses on multi‐hop communication among the patients on whose body, the body sensor nodes are installed, so that the data through the multi‐hop reach to the healthcare. The Relay Head (RH) node is used for forwarding the data to next RH and hence to sink. RH selection follows the same method of selection as that of CH. The experimental outcomes of NEEMA outperform the state‐of‐the‐art algorithms and prove to be highly beneficial for various WBAN applications.

A Novel Energy Efficient hybrid Meta‐heuristic Approach (NEEMA) for wireless body area network

Temporal sensitive heterogeneous graph neural network for news recommendation

A novel nest-based scheduling method for mobile wireless body area networks

An information cascade occurs when a person observes the actions of others and then engages in the same acts. Cascades may break out if a large population of nodes in the network get affected. The outbreaks of cascades will often bring influential events, which leads to an open research problem: how to accurately predict the cascading outbreaks in social networks? Although there have been some existing works on cascading outbreak prediction, they ignore the structure information of cascades. In this paper, we propose to use subcascades as features for cascade outbreak prediction. We use frequent sequential pattern mining to extract subcascades and then propose a max-margin based classifier to select at most B features for prediction. The proposed model is empirically evaluated on both synthetic and real-world networks. Experimental results demonstrate the effectiveness of the proposed model.

Mining subcascade features for cascade outbreak prediction in big networks

With the advent of web2.0 era, it has been becoming increasingly easy to create and share Internet content. Plenty of pictures are uploaded to the Internet every day. A primary challenge against traditional image retrieval technologies is how to help users quickly discover the images they need. Personalised image retrieval is a new trend in the field of image retrieval. It not only improves the accuracy of the existing retrieval systems, but also better meets the users’ needs. Personalised image retrieval and recommendation (PIRR) can be grouped into two main categories, content-based PIRR and collaborative filtering (CF)-based PIRR. This paper first summarises the development of image retrieval and introduces different image retrieval solutions. Then the key technologies of content-based PIRR are analysed from three aspects, user interest acquisition, user interest representation and personalised implementation. Different techniques are compared and analysed. Regarding CF-based PIRR, the user-based, item-based and model-based CF-based PIRR are introduced and compared. At the end of the paper, we compare and summarise content-based PIRR and CF-based PIRR.

A survey of personalised image retrieval and recommendation

s for Invited Talks Combinatorial Online Learning

Zhenyan Ji

Papers

Temporal sensitive heterogeneous graph neural network for news recommendation

A novel nest-based scheduling method for mobile wireless body area networks

Mining subcascade features for cascade outbreak prediction in big networks

A survey of personalised image retrieval and recommendation

Theoretical Computer Science