Research Article  Open Access
Combining Review Text Content and ReviewerItem Rating Matrix to Predict Review Rating
Abstract
Ecommerce develops rapidly. Learning and taking good advantage of the myriad reviews from online customers has become crucial to the success in this game, which calls for increasingly more accuracy in sentiment classification of these reviews. Therefore the finergrained review rating prediction is preferred over the rough binary sentiment classification. There are mainly two types of method in current review rating prediction. One includes methods based on review text content which focus almost exclusively on textual content and seldom relate to those reviewers and items remarked in other relevant reviews. The other one contains methods based on collaborative filtering which extract information from previous records in the revieweritem rating matrix, however, ignoring review textual content. Here we proposed a framework for review rating prediction which shows the effective combination of the two. Then we further proposed three specific methods under this framework. Experiments on two movie review datasets demonstrate that our review rating prediction framework has better performance than those previous methods.
1. Introduction
Web 2.0 and ecommerce give rise to the explosion of online reviews. In turn, intelligently learning the sentiment propensity and opinions of these reviews is exactly the key to success in current wave of ecommerce. Binary classification, or positivenegative classification, of these reviews has been quite common but it gradually fails to meet the requirement of accuracy [1]. For instance, which item will be selected out of several items that all belong to the positive category is, therefore, hard to predict. However, even the nuances between review rates can lead to great difference in their volume of sales. The review rating prediction research of [2] shows that consumers are often willing to pay 20% to even 99% more money to buy a product of 5star rate than that of 4star rate.
There are mainly two types of methods of current review rating prediction (RRP). The first one based on review text content adopts the perspective of natural language process. Researchers transform review text into feature vectors and then employ a multiclass classifier or a regression model to predict review rates [3–7]. It simply ignores the relationship between the costumers and the items. The second one based on collaborative filtering (CF) focuses on the standpoints of recommenders [8, 9]. Researchers employ the nearest neighbor methods [10, 11] or matrix factorization methods [12–17] to extract information from previous revieweritem rating matrix for review rating prediction. This type of method exploits no information from review text.
In order to include more information to accomplish finergrained review rating prediction, we proposed a framework combining review text content with previous revieweritem rating matrix. After that, we contrived three specific methods under this framework. Then we did some experiments on two movie review datasets to examine the efficiency of our framework and three methods. And the result shows that our methods under this framework by and large refine the performance of RRP and generate desirable results.
The outline of the paper is as follows. Section 2 introduces those related researches on RRP. RRP based on review text content and RRP based on CF are described, respectively, in Section 3. Section 4 presents our framework and the three methods under it. Experimental results on two movie review datasets are reported in Section 5. Finally, Section 6 concludes the paper and points out our future research direction.
2. Related Work
The research of sentiment classification is mainly divided into two aspects [18, 19]: one is positivenegative binary classification; the other is the finegrained RRP. As to binary classification of reviews, some of the creative methods for achieving this have been discussed in [20, 21]. In RRP, there are many research results proposed [3–17], but accuracy of RRP still cannot meet the real demands at present.
At present, there are mainly two ways of fulfilling finergrained RRP. One includes methods based on review text content (MBRTC), which mines information from review text content by discerning and quantifying a variety of text features and then employs the regression model to predict review rating [7, 22]. For example, Qu et al. [7] consider RRP as a feature engineering problem, and extracted various features, such as words, patterns, syntactic structures, and semantic topics, from the review text to improve the performance. Wang et al. [22] proposed a type of methods based on the content of review and weighting strong social relation of reviewers to predict review rating. To be specific, they predict review rating by incorporating the character of reviewer’s social relations, as regularization constraints, into contentbased methods. The main problem of MBRTC is that it mainly uses the review text content information and does not refer to revieweritems rating matrix information.
There are also some researches based on review text content taking into account characteristics of the items or the reviewers [5, 23]. Wang et al. [23] noticed that the score that a piece of review relates to cannot be fully determined by the review content itself, since review content is not an absolute metric of sentiment propensity. A tough reviewer may use tough words for all items, even items that he rates high. Different items have to meet different basic requirements. Simply analyzing the review content is not enough. Li et al. [5] proposed a method of incorporating reviewer and item information into review text content. They consider the personal characteristics of the reviewers when mining reviews content and use tensor factorization techniques to learn parameter of regression model and predict reviews rating. This method only considers the effect of reviewer and item on review text content and then uses review text content to predict review rating, which is indeed a method based on review text content.
The other one contains methods based on CF (MBCF). Those methods can be further divided into two categories. The first one dictates people to do similarity calculation to find the nearest neighbor reviewers or items to do prediction [11, 12]. The second one requires people to use the latent factor model to fulfill the matrix factorization. Several lowdimensional matrix factorization techniques are presented in [24–26]. Koren et al. [12–14] proposed several enhanced matrix factorization methods which can generate promising results by applying heterogeneous information to object functions. Koren [12] built a combined model by merging the matrix factorization and neighborhood models and improved accuracy of recommendation by extending the models to exploit both explicit and implicit feedback by the users. Koren [14] proposed a method to model the time changing behavior throughout the life span of the data and improved the performance of recommendation. In [27], researchers extended the matrix factorization objective function with the social network information of reviewers. In [28], Shi et al. proposed a contextaware movie recommendation algorithm based on joint matrix factorization (JMF). They jointly factorize the useritem matrix containing general movie ratings and other contextual movie similarity matrixes to integrate contextual information into the recommendation process.
Up to now, some researches combining ratings and text reviews have been applied to recommend system [29, 30]. For example, Cremonesi et al. proposed a hotel recommender algorithm (Interleave), which provides recommendations based on the text reviews and ratings [29]. Levi et al. proposed a recommender system that combines reviews and ratings to recommend hotels [30]. But as far as we know, the methods based on combining ratings and text reviews have not been applied in review rating prediction. Different from the existing methods focusing on recommend system, we focused on review rating prediction and proposed a general framework and three special methods based on review text content and revieweritem matrix.
Different from the above methods, we propose a framework combining MBRTC and MBCF to include more information to improve the accuracy of prediction. We also present three specific methods under this framework. Finally, the experiment results verify effectiveness of our proposed framework and methods.
3. RRP Based on Review Text Content and RRP Based on ReviewerItem Rating Matrix
For an online review site with items , reviews about the items, reviewers who have written the reviews, and review ratings which are corresponding to reviews , our goal is to predict rating of each review in . In this section, we introduce two existing types of methods in review rating prediction. One is based on review text content; the other one is based on collaborative filtering.
3.1. RRP Based on Review Text Content
Review text content is a very important information source for RRP. Current review text contentbased RRP methods mainly use vector space model (VSM) to express review text content and then use linear regression model to predict review rating. To be specific, there are four steps to take. Firstly, online review text content, which includes segmentations of terms, partofspeech tagging, and frequency statistics, should be preprocessed. Secondly, regarding words, phrases, and gram as features, people employ some feature selection methods to choose features that can perfectly express the review text content to compose the feature set . Thirdly, each online review in is expressed as an dimensional vector which is exactly an instantiated value of . Fourthly, the linear regression model dealing with those vectors of reviews is adopted to predict review rating. The linear regression model is described in
To work out the parameter vector , given training datasets with and available, least squares error loss is used to minimize the objective function:
Here, , the regularization term of parameter vector , is employed to avoid overfitting; is the regularization coefficient. To estimate the parameters , a simple stochastic gradient descent algorithm is adopted to solve the optimization problem. For each observed rating , we refer to the following updating rules to learn the parameters :
Here , and are the learning rates. After getting , given , we can apply to predict the review rating of each review in .
3.2. RRP Based on Collaborative Filtering
RRP now plays an essential role in recommend system. At present, RRP for recommend system mainly based on collaborative filtering involves two methods. One uses the nearest neighbor to predict and estimate the current object. The other one uses matrix factorization.
3.2.1. RRP Based on the Nearest Neighbor Model
RRP based on the nearest neighbor model includes the reviewerbased method and the itembased method. With the revieweritem rating matrix available, a typical reviewerbased approach is to predict a reviewer’s rating on a target item by aggregating the previous ratings on it from nearest reviewers. We can consequently formulate the predicted rating on item from reviewer as follows:
Here, is the set of nearest neighbor of reviewer and represents the similarity between reviewer and reviewer ; is review rating from reviewer on item .
To get the parameter , given training datasets with and available, we have to solve the optimization problem, that is, to minimize the square error loss function below:
Here is a regularization term of parameter aimed at avoiding overfitting and is regularization coefficient. Then, a simple stochastic gradient descent algorithm is adopted to solve the optimization problem. For each observed rating , we refer to the following updating rules to learn the parameters :
Here, and are the learning rates. After getting , we can apply to predict the review rating.
3.2.2. RRP Based on Matrix Factorization
Matrix factorization (MF) is one of the most popular methods in recommend system. The kernel of MF is to find a small number of latent features that might relate to the preferences of reviewer and use them to match observed ratings. A typical model associates each reviewer with a vector of reviewer factors and each item with a vector of item factors. The prediction is done through an inner product which is described by
In order to compute the two parameters and , we follow the least squares error loss principle to minimize the objective function:
Here and are the regularization terms of parameters and serving to avoid overfitting; is the regularization coefficient. In order to estimate the parameters and , a simple gradient descent algorithm was successfully applied to solve the optimization problem. For each observed rating , we used the following updating rules to acquire the parameters and :
Here, and are the learning rates. After getting and , we can apply to predict review rating.
4. RRP by Combining Review Text Content and ReviewerItem Rating Matrix
4.1. Problem Description
In order to illustrate the problem we study in this paper, a toy example about reviewers, items, review text content, and review rating is shown in Table 1. From the toy example, we can get three types of information: the useritem rating matrix, review text content with corresponding rating, and review text content without corresponding rating. The problem we study is how to effectively predict missing rating of each review in useritem rating matrix. In this section, we propose a new RRP framework combining revieweritem rating matrix (RIRM) with review text content (RTC). That is, we want to find a function .

4.2. General Framework
There are mainly two types of methods in existing RRP. One includes the methods based on review text content, which can be described as a function . The other one contains the methods based on collaborative filtering, which can be described as a function .
The above two types of methods use either review text content or revieweritem rating matrix, not having made full use of all the information available. Therefore, we proposed a RRP framework combining review text content and revieweritem rating matrix. The framework is described in
According to the general framework, we proposed three specific special RRP methods. In order to improve the performance of the three special RRP methods, we contrived a way to compute the parameters in the three special RRP methods.
4.3. Three Special RRP Methods
4.3.1. RRP Combining Linear Regression Model and Nearest Neighbor
We choose the linear regression model as and nearest neighbor to form . The special methods are described in
In order to get optimum parameters , , , given training datasets with reviewers who have written review ratings which are corresponding to reviews , we follow the least squares error loss principle to minimize the objective function:
Here and are regularization terms of those parameters aimed at avoiding overfitting; is a regularization coefficient. In order to estimate the parameters , , and , we firstly traverse from 0 to 1. Secondly, for each fixed in training dataset, we then adopt a simple stochastic gradient descent algorithm to solve the optimization problem. For each observed rating , we refer to the following updating rules:
Here , and are the learning rates. After getting , , and , given , we can apply to predict review rating of each review in .
4.3.2. RRP Combining Linear Regression Model and Matrix Factorization
Here, we choose linear regression model as and Matrix Factorization as . The special methods are described in
To acquire optimum parameters , , , and , given training datasets, we minimize the objective function according to the least square error loss principle:
Here , , and are regularization terms of those parameters serving to avoid overfitting; is a regularization coefficient. In order to estimate the parameters , , , and , we firstly traverse from 0 to 1. Secondly, for each fixed in training dataset, we adopt a simple stochastic gradient descent algorithm to solve the optimization problem. For each observed rating , we refer to the following updating rules:
Here , and are the learning rates. After learning , , , and , for , we can apply to predict the review ratings of it.
4.3.3. RRP by Combining Linear Regression Model, the Nearest Neighbor, and Matrix Factorization
In RRP framework, we choose linear regression model as and nearest neighbor and Matrix Factorization as . The special methods are described in
In order to get optimum parameters , , , , , and , given training datasets available, we follow the least square error loss principle to minimize the objective function:
Here , , , and are regularization terms of those parameters serving to avoid overfitting; is a regularization coefficient. In order to estimate the parameters , , , , , and , we firstly traverse from 0 to 1. Secondly, for each fixed in training datasets, we traverse from 0 to 1. Thirdly, for each fixed and in training datasets, we adopt a simple stochastic gradient descent algorithm to solve the optimization problem. For each observed rating , we refer to the following updating rules:
Here and are the learning rates. After learning , , , , , and , for , we can apply to predict the review ratings of it.
5. Evaluations
5.1. Datasets and Experimental Setup
In order to verify the performance of our proposed framework and methods, we performed several experiments on two datasets from the popular review site http://www.douban.com/. This website is a revieweropinion website where reviewers can read and write reviews on movies, music, and books and mark a rating from 1 star to 5 stars. We downloaded information of both movies and reviewers, movie reviews, and their ratings through the API of http://movie.douban.com/ community. The rough description of the two datasets is shown in Table 2.

In Table 2, revieweritem rating matrix density (RIRMD) is calculated by
To evaluate the overall performance of our framework and methods, we divide datasets into 10 parts at random. We do the experiment with taking 80% reviews for training and the remaining 20% reviews for test. We compare our framework and methods with methods based either on review text content or on revieweritem rating matrix through experiments on two different datasets. The six different methods are abbreviated as follows: MBLR: methods based on linear regression model. MBKNN: methods based on nearest neighbor. MBMF: methods based on matrix factorization. MBLR + MBKKN: our method combining linear regression model and nearest neighbor. MBLR + MBMF: our method combining linear regression model and matrix factorization. MBLR + MBKNN + MBMF: our method combining linear regression model, nearest neighbor, and matrix factorization.
There are mainly two factors influencing RRP in our framework. One is review text content information. The second is reviewitem matrix information. We did four experiments on two datasets to answer the following four questions:(1)How to set parameters and , and what are the effects of different parameters and on MAE and RMSE of RRP?(2)Can our framework and methods decrease MAE and RMSE of RRP?(3)Is the algorithm complexity of our methods higher than the three single methods?(4)What is the relationship between RIRMD and MAE and RMSE of RRP?
We use root mean square error (RMSE) and mean absolute error (MAE) as metric to evaluate performance of different RRP methods. RMSE and MAE are computed by where is the predicting rating by all kinds of methods, is the rating we have got from test datasets, and is the number of reviews in test datasets.
5.2. Performance Evaluation
5.2.1. Setting Parameters and in Our Frameworks
In our framework, there are two parameters that have to be set, namely, and . We perform 10fold cross validation in training datasets to get the optimum value of parameters and . Figure 1 shows how MAE and RMSE of MBLR + MBKKN change with parameter in two different training datasets. Figure 2 shows how MAE and RMSE of MBLR + MBMF change with parameter in two different training datasets. Figure 3 shows how MAE and RMSE of MBLR + MBKNN + MBMF change with parameter in two different training datasets.
(a)
(b)
(c)
(d)
From Figures 1, 2, and 3, we can see that MAE and RMSE of RRP in our methods change with different parameters and , which proved that parameters and play very important roles in our methods. So, we need to find optimum parameters and in our methods to improve the performance of RRP.
In Figure 1, when parameter = 0.01, MAE and RMSE of MBLR + MBKNN are the lowest. The reason is that revieweritems rating matrix of dataset 1 and dataset 2 is very sparse. It is well known that nearest neighbor method based on CF has poor performance on very sparse datasets. At the same time, linear regression model based on review text content has better performance than nearest neighbor method based on CF in dataset 1 and dataset 2.
In Figure 2, when parameter is set between 0.5 and 0.6, MAE and RMSE of MBLR + MBMF are the lowest. The reason is that MF method based on CF has slightly better performance than linear regression model based on review text content.
In Figure 3, we fix . When parameter is chosen between 0.5 and 0.6, MAE and RMSE of MBLR + MBKNN + MBMF are the lowest. The reason is that both MF method based on CF and linear regression model based on review text content have better performance than nearest neighbor method based on CF when revieweritems rating matrix of datasets is very sparse. At the same time, MF method based on CF has slightly better performance than linear regression model based on review text content.
Therefore, we set those parameters of our three methods according to the above results. The special parameters of the three different methods are shown in Table 3.

The MAE and RMSE of the three single methods are changed in dataset 1 and dataset 2, but the relative performance of the three single methods is basically unchanged in dataset 1 and dataset 2. For example, the gap between RMSE of the MBKNN and MBLR in dataset 1 and dataset 2 is basically identical (0.353). The gap between RMSE of the MBLR and MBMF in dataset 1 and dataset 2 is basically identical (0.035). So when we combining the three different single methods, we can preliminary choose parameter according to the relative performance of the three different single methods. The gap between RMSE of MBLR and MBMF is 10 times bigger than the gap between RMSE of MBKNN and MBLR.
From Figures 1, 2, and 3, we can see that the performance of our proposed methods is continuously better than single methods when parameter is chosen according to the relative performance of the three different single methods. At the same time, when parameter changed in range of the relative performance of the three different single methods, our proposed methods are continuously better than single methods. For example, we can see that parameter beta changes between 0.1 and 0.9; our methods obtain always better performance than MBKNN, MBLR, and MBMF in Figures 2 and 3. So, in real world applications, our methods, which are obtained by training with one dataset, can be applied to many different ones.
5.2.2. Effect of Different Methods to MAE and RMSE of RRP
In order to verify the performance of our proposed framework and methods, we compare our methods with three baseline methods in two different datasets. The experience results of those six methods on two datasets are presented in Table 4.
 
: dimension of VSM; : number of reviewer’s nearest neighbor; : number of latent factor in ratings matrix. 
In both two datasets, MAE and RMSE of MBLR + MBKNN are lower than that of MBLR and MBKNN; MAE and RMSE of MBLR + MBMF are lower than that of MBLR and MBMF; MAE and RMSE of MBLR + MBKNN + MBMF are lower than that of MBLR, MBKNN, and MBMF. Experimental results prove that combining text content information and reviewitem matrix information can enhance the performance of RRP. This is because both of the two different information sources are not redundant and can therefore play their own role. When we combine the two types of information: text content information and reviewitem matrix information, performance of RRP is improved in a certain extent.
For example, MBLR + MBKNN decreases 4.7% ((1.100798 − 1.049105)/1.100798) RMSE compared to MBLR in dataset 1. MBLR + MBMF decreases 5.73% ((1.065305 − 1.004237)/1.065305) RMSE compared to MBMF in dataset 1. MBLR + MBKNN + MBMF decreases 6.51% ((1.065305 − 0.995946)/1.065305) RMSE compared to MBMF in dataset 1. Although MBLR + MBKNN + MBMF only decreases 0.78% RMSE compared to MBLR + MBMF, both the two methods are our proposed methods. Compared to the three single methods, our proposed methods all decrease the RMSE of RRP and improve the performance of RRP.
5.2.3. Analyzing Complexity of Different Methods
Firstly, we compute the algorithm complexity of three simple approaches in Section 3. When computing the parameter of the objective function in the three methods, a simple stochastic gradient descent algorithm is adopted. According to the complexity of the stochastic gradient descent algorithm, we can get the algorithm complexity of the three simple approaches in Section 3. The algorithm complexity of those three methods on two datasets is presented in Table 4.
Then we compute the algorithm complexity of three special RRP approaches in Section 4. Similar to Section 3, when computing the parameter of the three special RRP methods, we also adopt a simple stochastic gradient descent algorithm to solve the optimization problem of minimizing the objective function. According to the complexity of the stochastic gradient descent algorithm and the computing course of the three special RRP methods, we can get the algorithm complexity of the three specific special RRP methods. The algorithm complexity of those three methods on two datasets is presented in Table 4.
From Table 4, we can see that the algorithm complexity of our proposed three methods in Section 4 is one order of magnitude with the algorithm complexity of the three simple methods in Section 3. When our methods obtain better results than the individual methods, the cost of our method is acceptable.
5.2.4. The Relations between Performance of RRP and RIRMD
In order to evaluate the effect that the revieweritem rating matrix density may have on review rating prediction, we experimented on two movie review datasets which are different in RIRMD. The experimental results are shown in Figures 4 and 5.
RIRMD of dataset 2 is denser than dataset 1. From Figures 4 and 5, we can see that MAE and RMSE values of dataset 2 are always lower than that of dataset 1 in our methods, which suggests that higher RIRMD always brings about lower MAE and RMSE. This is because higher RIRMD means more sufficient revieweritem matrix information. When we combine more sufficient revieweritem matrix information with review text content, we can get more accurate result statistically.
6. Conclusion
In this paper, we studied the previous methods of RRP and proposed a RRP framework combining review text content and revieweritem rating matrix to make full use of all information sources to improve the performance of prediction. Based on RRP framework, we further contrived three specific RRP methods. Our methods have significantly enhanced the performance of RRP, compared to methods based solely on review text content or collaborative filtering. In the future, we will further experiment on frameworks combining review text content and revieweritem matrix while employing the probability graph models.
Conflict of Interests
The authors declare that there is no conflict of interests regarding the publication of this paper.
Acknowledgments
The authors thank Lei Liu and Yuchong Li for helpful comments. This work was supported in part by National Key Technology Support Program (2014BAH41B00) and National Natural Science Foundation of China (U1405254, 61472092, 61402115, and 61271392).
References
 B. Pang, L. Lee, and S. Vaithyanathan, “Thumbs up?: sentiment classification using machine learning techniques,” in Proceedings of the ACL02 Conference on Empirical Methods in Natural Language Processing—Volume 10 (EMNLP '02), pp. 79–86, Association for Computational Linguistics, Stroudsburg, Pa, USA, July 2002. View at: Publisher Site  Google Scholar
 J. Horrigan, “Online shopping,” Pew Internet and American Life Project Report, Pew Research Center, 2008. View at: Google Scholar
 B. Pang and L. Lee, “Seeing stars: exploiting class relationships for sentiment categorization with respect to rating scales,” in Proceedings of the 43rd Annual Meeting of the Association for Computational Linguistics (ACL '05), pp. 115–124, June 2005. View at: Google Scholar
 S. Moghaddam and M. Ester, “ILDA: interdependent LDA model for learning latent aspects and their ratings from online product reviews,” in Proceedings of the 34th International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR '11), pp. 665–674, ACM, Beijing, China, July 2011. View at: Publisher Site  Google Scholar
 F. Li, N. Liu, H. Jin, K. Zhao, Q. Yang, and X. Zhu, “Incorporating reviewer and product information for review rating prediction,” in Proceedings of the 22nd International Joint Conference on Artificial Intelligence (IJCAI '11), vol. 3, pp. 1820–1825, AAAI Press, Barcelona, Spain, July 2011. View at: Publisher Site  Google Scholar
 G. Ganu, N. Elhadad, and A. Marian, “Beyond the stars: improving rating predictions using review text content,” in Proceedings of the 12th International Workshop on the Web and Databases (WebDB '09), pp. 1–6, 2009. View at: Google Scholar
 L. Qu, G. Ifrim, and G. Weikum, “The bagofopinions method for review rating prediction from sparse text patterns,” in Proceedings of the 23rd International Conference on Computational Linguistics (COLING '10), pp. 913–921, Association for Computational Linguistics, Beijing, China, August 2010. View at: Google Scholar
 Y. Shi, M. Larson, and A. Hanjalic, “Collaborative filtering beyond the useritem matrix: a survey of the state of the art and future challenges,” ACM Computing Surveys, vol. 47, no. 1, Article ID 2556270, pp. 3:1–3:45, 2014. View at: Publisher Site  Google Scholar
 H. Ma, “An experimental study on implicit social recommendation,” in Proceedings of the 36th International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR '13), pp. 73–82, ACM, New York, NY, USA, August 2013. View at: Publisher Site  Google Scholar
 M. Deshpande and G. Karypis, “Itembased topn recommendation algorithms,” ACM Transactions on Information Systems, vol. 22, no. 1, pp. 143–177, 2004. View at: Publisher Site  Google Scholar
 R. Jin, J. Y. Chai, and L. Si, “An automatic weighting scheme for collaborative filtering,” in Proceedings of the 27th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR '04), pp. 337–344, ACM, Sheffield, UK, July 2004. View at: Publisher Site  Google Scholar
 Y. Koren, “Factorization meets the neighborhood: a multifaceted collaborative filtering model,” in Proceedings of the 14th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD '08), pp. 426–434, August 2008. View at: Publisher Site  Google Scholar
 Y. Koren, R. Bell, and C. Volinsky, “Matrix factorization techniques for recommender systems,” Computer, vol. 42, no. 8, pp. 30–37, 2009. View at: Publisher Site  Google Scholar
 Y. Koren, “Collaborative filtering with temporal dynamics,” Communications of the ACM, vol. 53, no. 4, pp. 89–97, 2010. View at: Publisher Site  Google Scholar
 M. Kurucz, A. A. Benczúr, and B. Torma, “Methods for large scale svd with missing values,” in Proceedings of the KDD Cup and Workshop, San Jose, Calif, USA, August 2007. View at: Google Scholar
 R. Salakhutdinov and A. Mnih, “Bayesian probabilistic matrix factorization using markov chain Monte Carlo,” in Proceedings of the 25th International Conference on Machine Learning (ICML '08), pp. 880–887, ACM, Helsinki, Finland, July 2008. View at: Publisher Site  Google Scholar
 K. Yu, S. Zhu, J. Lafferty, and Y. Gong, “Fast nonparametric matrix factorization for largescale collaborative filtering,” in Proceedings of the 32nd Annual International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR '09), pp. 211–218, ACM, Boston, Mass, USA, July 2009. View at: Publisher Site  Google Scholar
 B. Pang and L. Lee, “Opinion mining and sentiment analysis,” Foundations and Trends in Information Retrieval, vol. 2, no. 12, pp. 1–135, 2008. View at: Publisher Site  Google Scholar
 B. Liu, Sentiment Analysis and Opinion Mining, Synthesis Lectures on Human Language Technologies, Morgan & Claypool Publishers, 2012. View at: Publisher Site
 B. Wang, Y. Min, Y. Huang et al., “Chinese reviews sentiment classification based on quantified sentiment lexicon and fuzzy set,” in Proceedings of the 3rd International Conference on Information Science and Technology (ICIST '13), pp. 678–680, IEEE, Yangzhou, China, March 2013. View at: Publisher Site  Google Scholar
 A. Bickerstaffe and I. Zukerman, “A hierarchical classifier applied to multiway sentiment detection,” in Proceedings of the 23rd International Conference on Computational Linguistics (COLING '10), pp. 62–70, Association for Computational Linguistics, Beijing, China, August 2010. View at: Google Scholar
 B. Wang, Y. Min, Y. Huang, X. Li, and F. Wu, “Review rating prediction based on the content and weighting strong social relation of reviewers,” in Proceedings of the International Workshop on Mining Unstructured Big Data Using Natural Language Processing (UnstructureNLP '13), pp. 23–30, ACM, Burlingame, Calif, USA, OctoberNovember 2013. View at: Publisher Site  Google Scholar
 H. Wang, Y. Lu, and C. Zhai, “Latent aspect rating analysis on review text data: a rating regression approach,” in Proceedings of the 16th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD '10), pp. 783–792, ACM, Washington, DC, USA, July 2010. View at: Publisher Site  Google Scholar
 J. D. M. Rennie and N. Srebro, “Fast maximum margin matrix factorization for collaborative prediction,” in Proceedings of the 22nd International Conference on Machine Learning (ICML '05), pp. 713–719, August 2005. View at: Publisher Site  Google Scholar
 A. Mnih and R. Salakhutdinov, “Probabilistic matrix factorization,” in Proceedings of the Advances in Neural Information Processing Systems (NIPS '07), pp. 1257–1264, Vancouver, Canada, 2007. View at: Google Scholar
 N. S. Nati and T. Jaakkola, “Weighted lowrank approximations,” in Proceedings of the 20th International Conference on Machine Learning, pp. 720–727, AAAI Press, Washington, DC, USA, August 2003. View at: Google Scholar
 H. Ma, D. Zhou, C. Liu, M. R. Lyu, and I. King, “Recommender systems with social regularization,” in Proceedings of the 4th ACM International Conference on Web Search and Data Mining (WSDM '11), pp. 287–296, ACM, Hong Kong, February 2011. View at: Publisher Site  Google Scholar
 Y. Shi, M. Larson, and A. Hanjalic, “Mining contextual movie similarity with matrix factorization for contextaware recommendation,” ACM Transactions on Intelligent Systems and Technology, vol. 4, no. 1, article 16, 2013. View at: Publisher Site  Google Scholar
 P. Cremonesi, F. Garzotto, and R. Turrin, “Usercentric vs. systemcentric evaluation of recommender systems,” in HumanComputer Interaction—INTERACT 2013, P. Kotz, G. Marsden, G. Lindgaard, J. Wesson, and M. Winckler, Eds., vol. 8119 of Lecture Notes in Computer Science, pp. 334–351, Springer, Berlin, Germany, 2013. View at: Publisher Site  Google Scholar
 A. Levi, O. Mokryn, C. Diot, and N. Taft, “Finding a needle in a haystack of reviews: cold start contextbased hotel recommender system demo,” in Proceedings of the 6th ACM Conference on Recommender Systems (RecSys '12), pp. 305–306, ACM, Dublin, Ireland, September 2012. View at: Publisher Site  Google Scholar
Copyright
Copyright © 2016 Bingkun Wang et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.