Over the last decade, social networking sites have become the most frequent way to connect online, which has led to the rise of underlying friend recommendation structure in social networks which suggests friends to users. Most existing friend recommendation frameworks, unfortunately, merely take into account the number of mutual friends, geo-location, mutual interests and other factors when recommending one person as a friend to another. Meanwhile, a number of recent research have demonstrated the value of deep learning and neural networks in the areas of recommendation systems, as well as recent improvements in the field of recommendation employing various deep learning variations. Thus, in this paper, a personalized friend recommendation system based on a hybrid model that combines link prediction (which is a widely used traditional method in most social media platforms and follows the friend-of-friend approach) with a neural network model for added accuracy and efficiency, is discussed.

Friend Recommendation System in a Social Network based on Link Prediction Framework using Deep Neural Network

Today, people frequently communicate through interactions and exchange knowledge over the social web in various formats. Social connections have been substantially improved by the emergence of social media platforms. Massive volumes of data have been generated by the expansion of social networks, and many people use them daily. Therefore, one of the current problems is to make it easier to find the appropriate friends for a particular user. Despite collaborative filtering’s huge success, accuracy and sparsity remain significant obstacles, particularly in the social networking sector, which has experienced astounding growth and has a large number of users. Social connections have been substantially improved by the emergence of social media platforms. In this work, a social and semantic-based collaborative filtering methodology is proposed for personalized recommendations in the context of social networking. A new hybrid collaborative filtering (HCoF) approach amalgamates the social and semantic suggestions. Two classification strategies are employed to enhance the performance of the recommendation to a high rate. Initially, the incremental K-means algorithm is applied to all users, and then the KNN algorithm for new users. The mean precision of 0.503 obtained by HCoF recommendation with semantic and social information results in an effective collaborative filtering enhancement strategy for friend recommendations in social networks. The evaluation’s findings showed that the proposed approach enhances recommendation accuracy while also resolving the sparsity and cold start issues.

/pdf/hcof-hybrid-collaborative-filtering-using-social-and-wembuf7t.pdf

HCoF: Hybrid Collaborative Filtering Using Social and Semantic Suggestions for Friend Recommendation

Machine learning algorithms are being explored and employed for various applications and have become the most sought topic of research in the modern era. It is a well known and accepted fact that a single machine learning algorithm cannot perform well for different applications. In this paper, effort is made to explore, design and evaluate eleven machine learning algorithms for four aerospace applications: O-ring failure prediction (classification and regression), Airfoil self noise prediction test (regression), Dynamics test (regression) and steel plate fault detection (classification). The performances of all the eleven algorithms were compared using the metric classification accuracy for classifiers and R2, RMSE metric for regressors. The algorithms were ranked based on their performance for all the above mentioned applications and the performance of proposed models are also compared with the results reported in the literature to conclude that the performance of proposed models are on par with the results reported in the literature. The proposed work can be easily extended to other Aerospace applications too.

Study and Evaluation of Machine Learning algorithms for Aerospace applications

Graph neural network (GNN) based recommendation models are observed to be more vulnerable against carefully-designed malicious records injected into the system, i.e., shilling attacks, which manipulate the recommendation to common users and therefore impair user trust. In this paper, we for the first time conduct a systematic study on the vulnerability of GNN based recommendation model against the shilling attack. With the aid of theoretical analysis, we attribute the root cause of the vulnerability to its neighborhood aggregation mechanism, which could make the negative impact of attacks propagate rapidly in the system. To restore the robustness of GNN based recommendation model, the key factor lies in detecting malicious records in the system and preventing the propagation of misinformation. To this end, we construct a user-user graph to capture the patterns of malicious behaviors and design a novel GNN based detector to identify fake users. Furthermore, we develop a data augmentation strategy and a joint learning paradigm to train the recommender model and the proposed detector. Extensive experiments on benchmark datasets validate the enhanced robustness of the proposed method in resisting various types of shilling attacks and identifying fake users, e.g., our proposed method fully mitigating the impact of popularity attacks on target items up to , and improving the accuracy of detecting fake users on the Gowalla dataset by .

Anti-FakeU: Defending Shilling Attacks on Graph Neural Network based Recommender Model

The study aims to present an architecture for a recommendation system based on user items that are transformed into narrow categories. In particular, to identify the movies a user will likely watch based on their favorite items. The recommendation system focuses on the shortest connections between item correlations. The degree of attention paid to user-group relationships provides another valuable piece of information obtained by joining the sub-groups. Various relationships have been used to reduce the data sparsity problem. We reformulate the existing data into several groups of items and users. As part of the calculations and containment of activities, we consider Pearson similarity, cosine similarity, Euclidean distance, the Gaussian distribution rule, matrix factorization, EM algorithm, and k-nearest neighbors (KNN). It is also demonstrated that the proposed methods could moderate possible recommendations from diverse perspectives.

/pdf/implementation-of-a-collaborative-recommendation-system-gzdtb7u8.pdf

Implementation of a Collaborative Recommendation System Based on Multi-Clustering

Friends Recommendation System identifies the behavior of users found in the dataset like user having number of followers, number of followings, common friends between followers and followings and provides the friend suggestions for the users they can follow. Recommendation system can also be used in other areas like recommending webpages to users in searching engines like Google, Explorer, Microsoft Edge. Recommending music in wynk, video recommendation in you tube, movie recommendation in amazon prime, recommendation of products to purchase in e-commerce applications like Flipkart, Amazon. Machine learning is used for providing recommendation on social networking application like Facebook, Instagram, Twitter Etc., In this Paper, recommendation of friends is done for Facebook. Similarity Coefficient calculations can be done using Jaccard Distance, Cosine Distance. Ranking Measures are done using Page Rank. Measuring the F1-Score and comparing the accuracy of different machine learning algorithms. These helps in finding which algorithm is more accurate in providing friends recommendations in social media recommendation system.

Friend's recommendation on social media using different algorithms of machine learning

There are already many software tools present and many application software's that being developed. These is because the technology needs to be updated according to the requirements. During these busy times, Developing and updating technologies is itself a big challenge. If any bugs occur in future or to keep track of features are added are working properly according to requirements, it needs some tools to keep track of it. For these things to happen, it requires some methodology to improve debug time like ETL (Event Trace Log) for multiple display features. The other things are that, code that is written today has to be written in such a way that it could be used for at least next 10 years. This will reduce the time of adding the same code simultaneously. Collecting the ETL in the encoded form helps in securing the sensitive information present in the trace log. Adding more testing codes in Unit Test Framework (UTF) find more bugs at early stages which will reduce the code cost. Adding MMIO (Memory Mapped I/O) verifier rules as the part of UTF helps in register verification (Simulates real Hardware context). Addition of MMIO rules for existing and new display features in separate xml file for different platforms makes it easy to identify in which platform feature failure in happening. All these methods help in code quality improvements for Intel Windows Graphics Driver.

Ruksar Parveen

Papers

Friend's recommendation on social media using different algorithms of machine learning

Code quality improvement for Intel Windows Graphics Driver