Image-Based Recommendations on Styles and Substitutes

McAuley, Julian; Targett, Christopher; Shi, Qinfeng; Hengel, Anton van den

doi:10.1145/2766462.2767755

Cited by 1,776 publications

(1,018 citation statements)

References 43 publications

Supporting

Mentioning

975

Contrasting

Order By: Relevance

“…We also compared the proposed models to some other state-of-the-art models. Our experiments used two datasets: the book dataset and the movie dataset, both from Amazon product dataset [38]. In the following, we first describe the datasets, the models used in comparison and the experiment configuration.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

A Multi-Modality Deep Network for Cold-Start Recommendation

Sun

Zhang

2018

BDCC

View full text Add to dashboard Cite

Collaborative filtering (CF) approaches, which provide recommendations based on ratings or purchase history, perform well for users and items with sufficient interactions. However, CF approaches suffer from the cold-start problem for users and items with few ratings. Hybrid recommender systems that combine collaborative filtering and content-based approaches have been proved as an effective way to alleviate the cold-start issue. Integrating contents from multiple heterogeneous data sources such as reviews and product images is challenging for two reasons. Firstly, mapping contents in different modalities from the original feature space to a joint lower-dimensional space is difficult since they have intrinsically different characteristics and statistical properties, such as sparse texts and dense images. Secondly, most algorithms only use content features as the prior knowledge to improve the estimation of user and item profiles but the ratings do not directly provide feedback to guide feature extraction. To tackle these challenges, we propose a tightly-coupled deep network model for fusing heterogeneous modalities, to avoid tedious feature extraction in specific domains, and to enable two-way information propagation from both content and rating information. Experiments on large-scale Amazon product data in book and movie domains demonstrate the effectiveness of the proposed model for cold-start recommendation.

show abstract

Section: Methodsmentioning

confidence: 99%

“…However, these models are typically simple neural network models and do not incorporate any content information. Recent work integrating deep learning with collaborative filtering mostly focuses on extracting content features from single modality such as texts [35][36][37] or images [38][39][40].…”

Section: Related Workmentioning

confidence: 99%

A Multi-Modality Deep Network for Cold-Start Recommendation

Sun

Zhang

2018

BDCC

View full text Add to dashboard Cite

show abstract

“…Fig. 6 provides a snapshot of Hadoop [32] interface which will be used for implementation of the proposed system. …”

Section: Methodsmentioning

confidence: 99%

A Framework for Development of Recommender System for Financial Data Analysis

Kanaujia¹,

Pandey²,

Rautaray³

2017

IJIEEB

View full text Add to dashboard Cite

Abstract-The huge amount of data is being created every day by various organisations and users all over the world. Structured, semi-structured and unstructured data is being created at a very rapid speed from heterogeneous sources like reviews, ratings, feedbacks, shopping details, etc., it is termed as Big Data. This data generated from different users share many common patterns which can be filtered and analysed to give some recommendation regarding the product, goods or services in which a user is interested. Recommendation systems are the software tools used to give suggestions to users on the basis of their requirements. Today no system is available for suggesting a person on how to use their money for saving, where to invest and how to manage expenditures. Few consulting systems are available which provide investment and saving tips but they are not much effective and are much complex. The presented paper proposed a collaborative filtering based recommender system for financial analysis based on Saving, Expenditure and Investment using Apache Hadoop and Apache Mahout. Many savings and investment consulting systems are available but no system provides effective and efficient recommendation regarding management and beneficial utilisation of salary. The advantage of proposed recommender system is that it provides better suggestion to a person for saving, expenditure and investment of their salary which in turns maximises their wealth. Due to enormous amount of data involved, Apache Hadoop framework is used for distributed processing. Collaborative filtering and Apache Mahout is used for analysing the data and implementation of the recommender system.

show abstract

“…Four of these datasets are obtained from Amazon [8], [9], three from MovieLens [10], [11], and one from Netflix [12]. These eight datasets used in our experiments (a) contain reliable timestamps (most of the ratings within each dataset have been entered in real rating time and not in a batch mode), (b) are up to date (published between 1998 and 2016), (c) are widely used as benchmarking datasets in CF research and (d) vary with respect to type of dataset (movies, music, videogames and books) and size (from 2MB, up to 4.7GB).…”

Section: Performance Evaluationmentioning

confidence: 99%

“…The proposed algorithm, as well as the two algorithms presented in [7], are based on the exploitation of timestamp information which is associated with ratings; hence in this work, we use the Amazon datasets [8], [9], the MovieLens datasets [10], [11] and the Netflix dataset [12], which include the ratings' timestamps. It is worth noting that the proposed algorithm can be combined with other techniques that have been proposed for either improving prediction accuracy in CF-based systems, including consideration of social network data (e.g.…”

Section: Introductionmentioning

confidence: 99%

Using Time Clusters for Following Users’ Shifts in Rating Practices

Margaris

Vassilakis

2017

CSIMQ

View full text Add to dashboard Cite

Abstract. Users that enter ratings for items follow different rating practices, in the sense that, when rating items, some users are more lenient, while others are stricter. This aspect is taken into account by the most widely used similarity metric in user-user collaborative filtering, namely, the Pearson Correlation, which adjusts each individual user rating by the mean value of the ratings entered by the specific user, when computing similarities. However, a user's rating practices change over time, i.e. a user could start as strict and subsequently become lenient or vice versa. In that sense, the practice of using a single mean value for adjusting users' ratings is inadequate, since it fails to follow such shifts in users' rating practices, leading to decreased rating prediction accuracy. In this work, we address this issue by using the concept of dynamic averages introduced earlier and we extend earlier work by (1) introducing the concept of rating time clusters and (2) presenting a novel algorithm for calculating dynamic user averages and exploiting them in useruser collaborative, filtering implementations. The proposed algorithm incorporates the aforementioned concept and is able to follow more successfully shifts in users' rating practices. It has been evaluated using numerous datasets, and has been found to introduce significant gains in rating prediction accuracy, while outperforming the dynamic average computation approaches that are presented earlier.

show abstract

Image-Based Recommendations on Styles and Substitutes

Cited by 1,776 publications

References 43 publications

A Multi-Modality Deep Network for Cold-Start Recommendation

A Multi-Modality Deep Network for Cold-Start Recommendation

A Framework for Development of Recommender System for Financial Data Analysis

Using Time Clusters for Following Users’ Shifts in Rating Practices

Contact Info

Product

Resources

About