Item Feature Vector Research Articles

Deep recommendation system based on knowledge graph and review text

With the explosive increase of information, recommendation system is applied in a variety of areas. However, the performance of recommendation system is limited due to issues such as data sparsity, cold starts and poor semantic understanding. In order to make full use of external information to assist recommendation, deeply mine the semantic information of review text and further improve the performance of recommendation system, a deep recommendation system based on knowledge graph and review text (Drs-kgrt) is proposed in this paper. In Drs-kgrt, knowledge graph, review text and the social records between users are used as auxiliary information to improve recommendation performance. Firstly, the review text is divided into user review text and item review text. BERT (Bidirectional Encoder Representation from Transformers) is used to accurately understand semantic information in user review text and the social records between users. The trust relationship between users and user preferences are fully mined to form user feature vectors. Secondly, BERT and knowledge graph entity recognition link technology are combined to extract item attribute feature entities and their associated entities. The fine-grained features of the items are analyzed to form item feature vectors. Thirdly, based on the scoring matrix, latent vectors of users and items are obtained by matrix decomposition. The deep features of users and items are generated based on user feature vectors, item feature vectors, latent vectors of users and items, the deep recommendation system is established to predict user scores for items. Finally, experiments are conducted on the Douban dataset and Amazon Movie Review dataset, the results show that the proposed algorithm can achieve better performance compared with other benchmark recommendation algorithms.

A Hybrid Solution For The Cold Start Problem In Recommendation

Abstract Recommender systems are becoming more and more significant in today’s digital world and in the modern economy. They make a substantial contribution to company operations by offering tailored advice and decreasing overwhelm. Collaborative filtering, being popular in the domain of recommendation, is used to offer recommendations to attract the target audience based on the feedback of people with comparable interests. This method has some limitations, such as a cold-start issue, which makes the system less effective in anticipating unknown objects. We provide a hybrid deep-learning-based strategy centered on a method to enrich user and item profiles to address the cold-start issue in the recommendation process using a collaborative filtering approach. We employ pretrained deep learning models to produce rich user and item feature vectors that aid in the creation of useful suggestions and handling of user and item cold-start issues. The creation of more precise and tailored similarity matrices is made possible by adding metadata to the extracted features of the user and item. The results of the experiment demonstrate that in terms of precision and rate coverage, the proposed method performs better than the baseline techniques.

Deep Learning and Embedding Based Latent Factor Model for Collaborative Recommender Systems

A collaborative recommender system based on a latent factor model has achieved significant success in the field of personalized recommender systems. However, the latent factor model suffers from sparsity problems. It is also limited in its ability to extract non-linear data features, resulting in poor recommendation performance. Inspired by the success of deep learning in different application areas, we incorporate deep learning into our proposed method to overcome the above problems. In this paper, we propose a dual deep learning and embedding-based latent factor model that considers dense user and item feature vectors. The model combines the existing deep learning and latent factor models to extract deep abstractions and non-linear feature representations of the data for rating prediction. The core idea is to map the dense user and item vectors generated by embedding techniques into dual, fully connected deep neural network architectures. In these two separate architectures, it learns the non-linear representation of the input data. The method then predicts the rating score by integrating the factors obtained from the two independent structures using the inner product. From the experimental result, we observe that the proposed model outperformed state-of-the-art existing models in real-world datasets (MovieLens ML-100K and ML-1M).

An Improved Web Service Recommendation Method based on Decomposition Machine Model

With the increasing number of Web services with similar functions on the Internet, traditional collaborative filtering service recommendation methods may encounter problems such as data sparseness, cold start, and poor scalability. To solve the above problems, this project proposes a new Web service recommendation method based on the decomposition machine model. The method decomposes the user trust relationship matrix and the product rating matrix while adding the geographic location information of the service, and transforms the correlation matrix of the calculated user feature vector and item feature vector into the same latent factor space by means of a decomposition machine. Optimize training model parameters to provide users with accurate prediction scores. The ultimate purpose of QoS prediction is to recommend high-quality services to users, improve the efficiency of users' discovery and selection of high-quality services, and ultimately promote the utilization of network Web services and promote service providers to release higher-quality services. Scientific significance, but also has better application value.

Deep transfer learning with multimodal embedding to tackle cold-start and sparsity issues in recommendation system.

Recommender systems (RSs) have become increasingly vital in the modern information era and connected economy. They play a key role in business operations by generating personalized suggestions and minimizing information overload. However, the performance of traditional RSs is limited by data sparseness and cold-start issues. Though deep learning-based recommender systems (DLRSs) are very popular, they underperform when considering rating matrices with sparse entries. Despite their performance improvements, DLRSs also suffer from data sparsity, cold start, serendipity, and generalizability issues. We propose a multistage model that uses multimodal data embedding and deep transfer learning for effective and personalized product recommendations, and is designed to overcome data sparsity and cold-start issues. The proposed model includes two phases. In the first—offline—phase, a deep learning technique is implemented to learn hidden features from a large image dataset (targeting new item cold start), and a multimodal data embedding is used to produce dense user feature and item feature vectors (targeting user cold start). This phase produces three different similarity matrices that are used as inputs for the second—online—phase to generate a list of top-n relevant items for a target user. We analyzed the accuracy and effectiveness of the proposed model against the existing baseline RSs using a Brazilian E-commerce dataset. The results show that our model scored 0.5882 for MAE and 0.4011 for RMSE which is lower than baseline RSs which indicates that the model achieved an improved accuracy and was able to minimize the typical cold start and data sparseness issues during the recommendation process.

Unexpected interest recommender system with graph neural network

AbstractTraditional recommender systems often face the filter bubble problem when they focus on recommending familiar items to users. The over-specialized recommended contents will make users bored. To solve this problem, researchers have proposed models that focus on unexpectedness, but these models all suffer from incomplete learning of features. To address this problem, we propose an unexpected interest recommender system with graph neural network (UIRS-GNN). First, we preprocess the input data with a graph convolutional network. It enriches user and item feature vectors by aggregating neighborhood information. Second, we transform the GRU and propose the attention-based long short-term gated recurrent unit network to learn user preferences hidden in historical behavior sequences. Then, we input the preprocessed feature vectors of users and items into the unexpected interest model, and solve the problem of insufficient feature information learning by aggregating neighborhood information. Furthermore, our model also alleviates data sparsity due to our deep learning feature information. Finally, empirical evaluations with several competitive baseline models on three real-world datasets reveal the superior performance of UIRS-GNN.

Quantization of Product using Collaborative Filtering Based on Cluster

Abstract: Because of strict response-time constraints, efficiency of top-k recommendation is crucial for real-world recommender systems. Locality sensitive hashing and index-based methods usually store both index data and item feature vectors in main memory, so they handle a limited number of items. Hashing-based recommendation methods enjoy low memory cost and fast retrieval of items, but suffer from large accuracy degradation. In this paper, we propose product Quantized Collaborative Filtering (pQCF) for better trade-off between efficiency and accuracy. pQCF decomposes a joint latent space of users and items into a Cartesian product of low-dimensional subspaces, and learns clustered representation within each subspace. A latent factor is then represented by a short code, which is composed of subspace cluster indexes. A user’s preference for an item can be efficiently calculated via table lookup. We then develop block coordinate descent for efficient optimization and reveal the learning of latent factors is seamlessly integrated with quantization. In this paper we also propose similarity method that has the ability to exploit multiple correlation structures between users who express their preferences for objects that are likely to have similar properties. For this we use a clustering method to find groups of similar objects. Index Terms: Product Quantization, Clustering, Product Search, Collaborative Filtering

DIAG: A Deep Interaction-Attribute-Generation model for user-generated item recommendation

Most existing recommendation methods assume that all the items are provided by separate producers rather than users. However, it could be inappropriate in some recommendation tasks since users may generate some items. Considering the user–item generation relation may benefit recommender systems that only use implicit user–item interactions. However, it may suffer from a dramatic imbalance. The number of user–item generation relations may be far smaller than the number of user–item interactions because each item is generated by at most one user. At the same time, this item can be interacted with by many users. To overcome the challenging imbalance issue, we propose a novel Deep Interaction-Attribute-Generation (DIAG) model. It integrates the user–item interaction relation, the user–item generation relation, and the item attribute information into one deep learning framework. The novelty lies in the design of a new item–item co-generation network for modeling the user–item generation information. Then, graph attention network is adopted to learn the item feature vectors from the user–item generations and the item attribute information by considering the adaptive impact of one item on its co-generated items. Extensive experiments conducted on two real-world datasets confirm the superiority of the DIAG method.

An Web Service Recommendation Method Based on Location and Decomposition Machine Model

With the increasing number of Web services with similar functions on the Internet, traditional collaborative filtering service recommendation methods may encounter problems such as data sparseness, cold start, and poor scalability. To solve the above problems, this project proposes a new Web service recommendation method based on the decomposition machine model. The method decomposes the user trust relationship matrix and the product rating matrix while adding the geographic location information of the service, and transforms the correlation matrix of the calculated user feature vector and item feature vector into the same latent factor space by means of a decomposition machine. Optimize training model parameters to provide users with accurate prediction scores. The ultimate purpose of QoS prediction is to recommend high-quality services to users, improve the efficiency of users' discovery and selection of high-quality services, and ultimately promote the utilization of network Web services and promote service providers to release higher-quality services. Scientific significance, but also has better application value.

Neural Explicit Factor Model Based on Item Features for Recommendation Systems

In recent years, recommendation systems based on collaborative filtering (CF) have achieved a high performance. Most of the existing recommendation systems use similarity measures to determine the suitability of items for users based on latent factor models (LFM). However, these recommendation systems reduce the explainability of recommendations and hide the reasons for recommending specific items. As a result, users tend to distrust the recommendation results. To address this problem, we propose the neural explicit factor model (NEFM). Based on the user–item rating matrix, we propose adding both user–feature attention matrix and an item–feature quality matrix to improve the explainability of user and item vectors. In addition, a feedforward neural network and a one-dimensional convolutional neural network extract features from user, item and the item–feature vector. Finally, a prediction layer performs the inner product of user data, item data, and item features. Experiments on the MovieLens and Yahoo Movies datasets validate the proposed model, and comparisons with similar recommendation models show the higher accuracy and explainability of our proposal.

Score Prediction Algorithm Combining Deep Learning and Matrix Factorization in Sensor Cloud Systems

In this era of exponential growth in the scale of data, information overload has become an urgent problem, and the use of increasingly flexible sensor cloud systems (SCS) for data collection has become a mainstream trend. Recommendation algorithms can search massive data sets to uncover information that meets the needs of users based on their interests. To improve the accuracy of recommendation scoring, this article proposes a score prediction algorithm that combines deep learning and matrix factorization. To address the problem of sparse scoring data, our study employs a sensor cloud system to collect data information, preprocesses the collected information, and then uses a deep learning model combined with explicit and implicit feedback to generate recommendations. The proposed algorithm, MF-NeuRec, combines fusion matrix decomposition and the NeuRec model score prediction algorithm. The algorithm employs user-based and item-based NeuRec algorithms to extract the feature vectors of users and items under implicit feedback data. The obtained user and item feature vectors are integrated in a certain ratio through the use of matrix decomposition under the display feedback data. The user and item feature vectors obtained by the algorithm are merged and analyzed to predict how users will rate items. Experiments demonstrate that the algorithm can improve the accuracy of recommendations.

Analysis and Research on Increased Probability Matrix Factorization Techniques in Collaborative Filtering

The matrix factorization algorithms such as the matrix factorization technique (MF), singular value decomposition (SVD) and the probability matrix factorization (PMF) and so on, are summarized and compared. Based on the above research work, a kind of improved probability matrix factorization algorithm called MPMF is proposed in this paper. MPMF determines the optimal value of dimension D of both the user feature vector and the item feature vector through experiments. The complexity of the algorithm scales linearly with the number of observations, which can be applied to massive data and has very good scalability. Experimental results show that MPMF can not only achieve higher recommendation accuracy, but also improve the efficiency of the algorithm in sparse and unbalanced data sets compared with other related algorithms.

Personalized Recommendation Algorithm Based on Deep Neural Network and Weighted Implicit Feedback

In singular value decomposition++(SVD++),inner product of user and item feature vector is regarded as user′s rating of items.However,inner product cannot capture the high-order nonlinear relationship between the user and the item.In addition,the contribution of different interactive items cannot be distinguished when user′s implicit feedback is incorporated in SVD++.A recommendation algorithm based on deep neural network and weighted implicit feedback is proposed to solve the two problems.Deep neural network is adopted to model the relationship between the user and the object and attention mechanism is utilized to calculate the weight of historical interactive items in modeling user′s implicit feedback.Experiments on public datasets verify the effectiveness of the proposed algorithm.

Product Quantized Collaborative Filtering

Because of strict response-time constraints, efficiency of top-k recommendation is crucial for real-world recommender systems. Locality sensitive hashing and index-based methods usually store both index data and item feature vectors in main memory, so they handle a limited number of items. Hashing-based recommendation methods enjoy low memory cost and fast retrieval of items, but suffer from large accuracy degradation. In this paper, we propose product Quantized Collaborative Filtering (pQCF) for better trade-off between efficiency and accuracy. pQCF decomposes a joint latent space of users and items into a Cartesian product of low-dimensional subspaces, and learns clustered representation within each subspace. A latent factor is then represented by a short code, which is composed of subspace cluster indexes. A user's preference for an item can be efficiently calculated via table lookup. We then develop block coordinate descent for efficient optimization and reveal the learning of latent factors is seamlessly integrated with quantization. We further investigate an asymmetric pQCF, dubbed as QCF, where user latent factors are not quantized and shared across different subspaces. The extensive experiments with 6 real-world datasets show that pQCF significantly outperforms the state-of-the-art hashing-based CF and QCF increases recommendation accuracy compared to pQCF.

Attribute mapping and autoencoder neural network based matrix factorization initialization for recommendation systems

The recommendation algorithm is attracting increasing attention in analyzing big data. Matrix factorization (MF) is one of the recommendation methods and Singular Value Decomposition (SVD) is the most popular matrix factorization method. However, the existing SVD methods usually initialize user and item feature randomly, not fully utilize the information of the data, so require plenty of experiments to determine feature matrix dimension, with low convergence efficiency and low accuracy. This paper presents a hybrid initialization method based on attribute mapping and autoencoder neural network to solve these problems, which consists of three parts: (1) use the number of item attribute types to determine feature matrix dimension in order to avoid multiple experiments to select the optimal dimension value; (2) use items’ attributes to initialize the item feature matrix in SVD＋＋, and use an attribute mapping mechanism to get an item feature vector by fitting the rating matrix to accelerate the convergence; (3) adopt the autoencoder neural network to reduce feature dimension and obtain item latent features for initializing SVD＋＋. The experimental results show that our methods achieve better performance than SVD＋＋ random initialization and also be adopted to other matrix factorization methods.

Constrained User Exposure Matrix Factorization in Recommendation System

Due to the massive items and users limited scope, many existing recommendation systems suffer a more and more serious sparse problem on rating matrix. Most of the recommendation systems employ the content-based method or the collaborative filter model to abbreviate the sparse problem. While in our paper, based on the assumption that users who have rated similar sets of items are likely to have similar preference information, we present a constrained user exposure matrix factorization collaborative filter model and apply it to deal with the matrix sparsity problem, which it makes a better recommendation to users who have very few ratings. In detail, for the sake of capturing the effect of user having rated a particular item, we introduce a constrained matrix to user latent vector space. This effect will have an impact on the prior mean of user latent feature vector. After that, using the user feature vector and item feature vector to estimate the values of unrated items. The experimental results show that the proposed model gets higher prediction precision than the state-of-the-art models in providing recommendation for user having very few ratings.

A Novel Recommendation System to Match College Events and Groups to Students

With the recent increase in data online, discovering meaningful opportunities can be time-consuming and complicated for many individuals. To overcome this data overload challenge, we present a novel text-content-based recommender system as a valuable tool to predict user interests. To that end, we develop a specific procedure to create user models and item feature-vectors, where items are described in free text. The user model is generated by soliciting from a user a few keywords and expanding those keywords into a list of weighted near-synonyms. The item feature-vectors are generated from the textual descriptions of the items, using modified tf-idf values of the users’ keywords and their near-synonyms. Once the users are modeled and the items are abstracted into feature vectors, the system returns the maximum-similarity items as recommendations to that user. Our experimental evaluation shows that our method of creating the user models and item feature-vectors resulted in higher precision and accuracy in comparison to well-known feature-vector-generating methods like Glove and Word2Vec. It also shows that stemming and the use of a modified version of tf-idf increase the accuracy and precision by 2% and 3%, respectively, compared to non-stemming and the standard tf-idf definition. Moreover, the evaluation results show that updating the user model from usage histories improves the precision and accuracy of the system. This recommender system has been developed as part of the Agnes application, which runs on iOS and Android platforms and is accessible through the Agnes website.

Learning Vector-space Representations of Items for Recommendations Using Word Embedding Models

We present a method of generating item recommendations by learning item feature vector embeddings. Our work is analogous to approaches like Word2Vec or Glove used to generate a good vector representation of words in a natural language corpus. We treat the items that a user interacted with as analogous to words and the string of items interacted with in a session as sentences. Our embedding generates semantically related clusters and the item vectors generated can be used to compute item similarity which can be used to drive product recommendations. Our method also allows us to use the feature vectors in other machine learning systems. We validate our method on the MovieLens dataset.

Application and Research of Improved Probability Matrix Factorization Techniques in Collaborative Filtering

The matrix factorization algorithms such as the matrix factorization technique (MF), singular value decomposition (SVD) and the probability matrix factorization (PMF) and so on, are summarized and compared. Based on the above research work, a kind of improved probability matrix factorization algorithm called MPMF is proposed in this paper. MPMF determines the optimal value of dimension D of both the user feature vector and the item feature vector through experiments. The complexity of the algorithm scales linearly with the number of observations, which can be applied to massive data and has very good scalability. Experimental results show that MPMF can not only achieve higher recommendation accuracy, but also improve the efficiency of the algorithm in sparse and unbalanced data sets compared with other related algorithms.

적응형 군집화 기반 확장 용이한 협업 필터링 기법

적응형 군집화 기반 확장 용이한 협업 필터링 기법