A standard model for Recommender Systems is the Matrix Completion setting: given partially known matrix of ratings given by users (rows) to items (columns), infer the unknown ratings. In the last decades, few attempts where done to handle that objective with Neural Networks, but recently an architecture based on Autoencoders proved to be a promising approach. In current paper, we enhanced that architecture (i) by using a loss function adapted to input data with missing values, and (ii) by incorporating side information. The experiments demonstrate that while side information only slightly improve the test error averaged on all users/items, it has more impact on cold users/items.

https://hal.inria.fr/hal-01336912/document

Hybrid Recommender System based on Autoencoders

Communication protocols for wireless sensor networks: A survey and comparison

Wireless sensor network (WSN) is a group of small power-constrained nodes that sense data and communicate it to the base station (BS). These nodes cover a vast region of interest (ROI) for several purposes according to the application need. The first challenge encountered in WSNs is how to cover the ROI perfectly and send the monitored data to the BS. Although the energy introduced during setup phase and the violation of energy fairness constraint of dynamic routing topologies, they achieve high network performance in terms of coverage and connectivity. In this paper, we categorize the applications of WSN based on different aspects to show the major protocol design issues. Thus, the energy efficiency of the recent proactive routing protocols is studied from different angles. The energy overhead and energy fairness of each protocol were carefully analyzed. The most energy efficient routing protocols for homogeneous proactive networks were studied and compared to highlight the research challenges and existing problems in this area. The results proved that energy overhead and route selection are the most effective aspects of network lifetime and network efficiency.

Survey on Wireless Sensor Network Applications and Energy Efficient Routing Protocols

Energy-efficient routing protocols for solving energy hole problem in wireless sensor networks

Energy efficient heterogeneous DEEC protocol for enhancing lifetime in WSNs

Recommender systems have shown a lot of awareness in the past decade. Due to their great business value, recommender systems have also been successfully deployed in business, such as product recommendation at flipkart, HomeShop18, and music recommendation at Last.fm, Pandora, and movie recommendation at Flixstreet, MovieLens, and Jinni. In the past few years, the incredible growth of Web 2.0 web sites and applications constitute new challenges for Traditional recommender systems. Traditional recommender systems always ignore social interaction among users. But in our real life, when we are asking our friends or looking opinions, reviews for recommendations of Mobile or heart touching music, movies, electronic gadgets, restaurant, book, games, software Apps, we are actually using social information for recommendations. In this paper social popularity factor are incorporated in SVD++ factorization method as implicit feedback to improve accuracy and scalability of recommendations.

/pdf/social-popularity-based-svd-recommender-system-30ubieyh9s.pdf

Social Popularity based SVD++ Recommender System

In this paper, a novel energy-efficient clustering protocol (NEECP) for increasing the network lifetime in wireless sensor networks is proposed. This technique selects the cluster heads in an effective way with an adjustable sensing range and performs data aggregation using chaining approach. It also avoids transmission of redundant data by using a redundancy check function for improving the network lifetime. It is implemented by considering the data with aggregation and without aggregation. The NEECP without aggregation increases the network lifetime by 59.76 and 7.17% as compared with the hybrid energy-efficient distributed (HEED) and intra-balanced low-energy adaptive clustering hierarchy (IBLEACH), respectively. It increases the network lifetime by 122.92 and 49.53% over the HEED and IBLEACH, respectively, while considering data aggregation.

https://ietresearch.onlinelibrary.wiley.com/doi/pdf/10.1049/iet-wss.2015.0017

NEECP: Novel energy-efficient clustering protocol for prolonging lifetime of WSNs

There have been discussed various data hiding methods which can embed the secret data in an image. However, the image gets destroyed after extraction of the secret data at the receiving end. Thus, there exists a need to have data hiding methods which can recover the image after extraction of the secret data. Such types of schemes are called reversible data hiding schemes which are commonly used in sensitive military, legal, and medical applications. The existing reversible data hiding schemes either provide good hiding capacity but low quality stego-image or good quality stego-image but poor data hiding capacity because the stego-image quality and the hiding capacity are diametrically related metrics. To address this problem, we propose a high capacity reversible data hiding scheme using recovery strategy. It hides the secret data into a cover image in two phases. In the first phase, the cover image is scanned in a specific scan order andconstructs a location map in which even valued pixels are denoted by ‘1’ and odd valued pixels by ‘0’. It embeds the secret data into every pixel of the image by changing its value at most by 1. The second phase repeats the process of the first phase embedding so that some of the pixels can berecovered to its original form as well as the secret data can further be embedded. Experimentally, it is proved that the proposed scheme provides good quality of stego-image and having the high data hiding capacity at the same time. Further, it is able to maintain the image quality even when the secret data is embedded in layers.

Recovery based high capacity reversible data hiding scheme using even-odd embedding

In this paper, we propose a new Absolute Moment Block Truncation Coding (AMBTC) based data hiding scheme using hamming distance and pixel value differencing methods. The proposed method firstly pre-processes the original image using a smoothing filter so that its quality can be maintained after AMBTC compression. It then applies AMBTC compression on the processed image and uses two thresholds to categorize the image blocks into three categories namely Smooth, Less_Complex, and Highly_Complex blocks so that the complex block can also be used to embed the secret data. The secret data is then embedded into the smooth blocks using simple replacement strategy. Further, the proposed method embeds 8 bits of secret data into the bit plane of Less_Complex blocks using hamming distance calculation and few bits into the quantization levels of Highly_Complex blocks using pixel value differencing (PVD) method to increase the data hiding capacity without having any major impact on image quality. Thus, the proposed method additionally embeds on average 6 more bits in complex blocks. Experimental results also prove that the proposed method is superior to the contemporary AMBTC based data hiding methods in term of data hiding capacity and has comparable image quality.

Rajeev Kumar

Papers

Energy efficient heterogeneous DEEC protocol for enhancing lifetime in WSNs

Social Popularity based SVD++ Recommender System

NEECP: Novel energy-efficient clustering protocol for prolonging lifetime of WSNs

Recovery based high capacity reversible data hiding scheme using even-odd embedding

Enhanced AMBTC based data hiding method using hamming distance and pixel value differencing