Iterative Compilation Optimization Based on Metric Learning and Collaborative Filtering

Pass selection and phase ordering are two critical compiler auto-tuning problems. Traditional heuristic methods cannot effectively address these NP-hard problems especially given the increasing number of compiler passes and diverse hardware architectures. Recent research efforts have attempted to address these problems through machine learning. However, the large search space of candidate pass sequences, the large numbers of redundant and irrelevant features, and the lack of training program instances make it difficult to learn models well. Several methods have tried to use expert knowledge to simplify the problems, such as using only the compiler passes or subsequences in the standard levels (e.g., -O1, -O2, and -O3) provided by compiler designers. However, these methods ignore other useful compiler passes that are not contained in the standard levels. Principal component analysis (PCA) and exploratory factor analysis (EFA) have been utilized to reduce the redundancy of feature data. However, these unsupervised methods retain all the information irrelevant to the performance of compilation optimization, which may mislead the subsequent model learning. To solve these problems, we propose a compiler pass selection and phase ordering approach, called Iterative Compilation based on Metric learning and Collaborative filtering (ICMC) . First, we propose a data-driven method to construct pass subsequences according to the observed collaborative interactions and dependency among passes on a given program set. Therefore, we can make use of all available compiler passes and prune the search space. Then, a supervised metric learning method is utilized to retain useful feature information for compilation optimization while removing both the irrelevant and the redundant information. Based on the learned similarity metric, a neighborhood-based collaborative filtering method is employed to iteratively recommend a few superior compiler passes for each target program. Last, an iterative data enhancement method is designed to alleviate the problem of lacking training program instances and to enhance the performance of iterative pass recommendations. The experimental results using the LLVM compiler on all 32 cBench programs show the following: (1) ICMC significantly outperforms several state-of-the-art compiler phase ordering methods, (2) it performs the same or better than the standard level -O3 on all the test programs, and (3) it can reach an average performance speedup of 1.20 (up to 1.46) compared with the standard level -O3.

Hybrid Variational Autoencoder for Recommender Systems

E-commerce platforms heavily rely on automatic personalized recommender systems, e.g., collaborative filtering models, to improve customer experience. Some hybrid models have been proposed recently to address the deficiency of existing models. However, their performances drop significantly when the dataset is sparse. Most of the recent works failed to fully address this shortcoming. At most, some of them only tried to alleviate the problem by considering either user side or item side content information. In this article, we propose a novel recommender model called Hybrid Variational Autoencoder (HVAE) to improve the performance on sparse datasets. Different from the existing approaches, we encode both user and item information into a latent space for semantic relevance measurement. In parallel, we utilize collaborative filtering to find the implicit factors of users and items, and combine their outputs to deliver a hybrid solution. In addition, we compare the performance of Gaussian distribution and multinomial distribution in learning the representations of the textual data. Our experiment results show that HVAE is able to significantly outperform state-of-the-art models with robust performance.

The Footprint of Factorization Models and Their Applications in Collaborative Filtering

Factorization models have been successfully applied to the recommendation problems and have significant impact to both academia and industries in the field of Collaborative Filtering ( CF ). However, the intermediate data generated in factorization models’ decision making process (or training process , footprint ) have been overlooked even though they may provide rich information to further improve recommendations. In this article, we introduce the concept of Convergence Pattern, which records how ratings are learned step-by-step in factorization models in the field of CF. We show that the concept of Convergence Patternexists in both the model perspective (e.g., classical Matrix Factorization ( MF ) and deep-learning factorization) and the training (learning) perspective (e.g., stochastic gradient descent ( SGD ), alternating least squares ( ALS ), and Markov Chain Monte Carlo ( MCMC )). By utilizing the Convergence Pattern, we propose a prediction model to estimate the prediction reliability of missing ratings and then improve the quality of recommendations. Two applications have been investigated: (1) how to evaluate the reliability of predicted missing ratings and thus recommend those ratings with high reliability. (2) How to explore the estimated reliability to adjust the predicted ratings to further improve the predication accuracy. Extensive experiments have been conducted on several benchmark datasets on three recommendation tasks: decision-aware recommendation, rating predicted, and Top- N recommendation. The experiment results have verified the effectiveness of the proposed methods in various aspects.

Improving collaborative filtering using lexicon-based sentiment analysis

<span>Since data is available increasingly on the Internet, efforts are needed to develop and improve recommender systems to produce a list of possible favorite items. In this paper, we expand our work to enhance the accuracy of Arabic collaborative filtering by applying sentiment analysis to user reviews, we also addressed major problems of the current work by applying effective techniques to handle the scalability and sparsity problems. The proposed approach consists of two phases: the sentiment analysis and the recommendation phase. The sentiment analysis phase estimates sentiment scores using a special lexicon for the Arabic dataset. The item-based and singular value decomposition-based collaborative filtering are used in the second phase. Overall, our proposed approach improves the experiments’ results by reducing average of mean absolute and root mean squared errors using a large Arabic dataset consisting of 63,000 book reviews.</span>

Implementation of a personalized food recommendation system based on collaborative filtering and knapsack method

Food recommendation system is one of the most interesting recommendation problems since it provides data for decision-making to users on selection of foods that meets individual preference of each user. Personalized recommender system has been used to recommend foods or menus to respond to requirements and restrictions of each user in a better way. This research study aimed to develop a personalized healthy food recommendation system based on collaborative filtering and knapsack method. Assessment results found that users were satisfied with the personalized healthy food recommendation system based on collaborative filtering and knapsack problem algorithm which included ability of operating system, screen design, and efficiency of operating system. The average satisfaction score overall was 4.20 implying that users had an excellent level of satisfaction.

Collaborative Reflection-Augmented Autoencoder Network for Recommender Systems

As the deep learning techniques have expanded to real-world recommendation tasks, many deep neural network based Collaborative Filtering (CF) models have been developed to project user-item interactions into latent feature space, based on various neural architectures, such as multi-layer perceptron, autoencoder, and graph neural networks. However, the majority of existing collaborative filtering systems are not well designed to handle missing data. Particularly, in order to inject the negative signals in the training phase, these solutions largely rely on negative sampling from unobserved user-item interactions and simply treating them as negative instances, which brings the recommendation performance degradation. To address the issues, we develop a C ollaborative R eflection-Augmented A utoencoder N etwork (CRANet), that is capable of exploring transferable knowledge from observed and unobserved user-item interactions. The network architecture of CRANet is formed of an integrative structure with a reflective receptor network and an information fusion autoencoder module, which endows our recommendation framework with the ability of encoding implicit user’s pairwise preference on both interacted and non-interacted items. Additionally, a parametric regularization-based tied-weight scheme is designed to perform robust joint training of the two-stage CRANetmodel. We finally experimentally validate CRANeton four diverse benchmark datasets corresponding to two recommendation tasks, to show that debiasing the negative signals of user-item interactions improves the performance as compared to various state-of-the-art recommendation techniques. Our source code is available at https://github.com/akaxlh/CRANet.

REKOMENDASI BUKU PERPUSTAKAAN KAMPUS DENGAN METODE ITEM-BASED COLLABORATIVE FILTERING

Artikel ini menawarkan alternatif solusi atas banyaknya jumlah buku yang terdapat dalam perpustakaan sehingga membuat beberapa mahasiswa kesulitan dalam menentukan pilihan mengenai buku yang tepat sesuai dengan ketertarikan mahasiswa. Metode rekomendasi yang akurat bisa menjadi sebuah solusi untuk mengatasi permasalahan tersebut. Metode item-based collaborative filtering merupakan metode yang memberikan prediksi sebuah item kepada pengguna berdasarkan ketertarikan dan opini dari pengguna lain. Penelitian ini menggunakan metode item-based collaborative filtering yang diterapkan pada rekomendasi untuk memberikan rekomendasi buku kepada mahasiswa. Data yang digunakan pada penelitian ini adalah data buku dan pengunjung dari tahun 2016 hingga 2019. Metode item-based collaborative bekerja dengan mencari nilai kemiripan suatu item yang belum pernah diberikan rating dengan item yang telah diberi rating menggunakan persamaan cosine similarity. Hasil perhitungan kemiripan antar item digunakan pada perhitungan prediksi rating menggunakan persamaan Weighted sum yang nilai prediksinya akan dijadikan rekomendasi kepada mahasiswa. Hasil penelitian menunjukkan bahwa metode item-based collaborative filtering mampu memberikan rekomendasi kepada mahasiswa dengan persentase pengujian akurasi MAPE sebesar 22% dan pengujian akurasi MAE sebesar 0.568.

Analysis of Brand Visual Design Based on Collaborative Filtering Algorithm

With the continuous development of China’s social economy, the competitiveness of brand market is gradually increasing. In order to improve their own level in brand building, major enterprises gradually explore and study visual communication design. Brand visual design has also received more and more attention. Building a complete and rich visual design system can improve the brand level and attract users to consume. Based on the abovementioned situation, this paper proposes to use collaborative filtering algorithm to analyze and study brand visual design. Firstly, a solution is proposed to solve the problem of low accuracy of general recommendation algorithm in brand goods. Collaborative filtering algorithm is used to analyze the visual communication design process of enterprise brand. Research on personalized image design according to consumers’ trust and recognition of brand design is conducted. In traditional craft brand visual design, we mainly study the impact of image design on consumer behavior. The brand loyalty model is used to predict and analyze the visual design effect. Also, the user’s evaluation coefficient is taken as the expression of brand visual design recognition. Finally, the collaborative filtering algorithm is optimized to improve the consumer similarity based on the original algorithm. The results show that the brand visual design using collaborative filtering algorithm can help enterprises obtain greater benefits in their own brand construction. It provides effective data help in the development of traditional craft brands.