Deep Cross-Domain Fashion Recommendation

Jaradat, Shatha

doi:10.1145/3109859.3109861

Cited by 35 publications

(17 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There are also other approaches which concentrate on fashion recommendation [2,7,8,18]. Hu et al [6] proposed a functional tensor factorization method to recommend outfits to users by learning from the interactions between user and fashion items.…”

Section: Fashion Recommendationmentioning

confidence: 99%

Enhancing Fashion Recommendation with Visual Compatibility Relationship

Yin

et al. 2019

The World Wide Web Conference

View full text Add to dashboard Cite

With the increasing of online shopping services, fashion recommendation plays an important role in daily online shopping scenes. A lot of recommender systems have been developed with visual information. However, few works take into account compatibility relationship when they are generating recommendations. The challenge is that fashion concept is often subtle and subjective for different customers. In this paper, we propose a fashion compatibility knowledge learning method that incorporates visual compatibility relationships as well as style information. We also propose a fashion recommendation method with domain adaptation strategy to alleviate the distribution gap between the items in target domain and the items of external compatible outfits. Our results indicate that the proposed method is capable of learning visual compatibility knowledge and outperforms all the baselines. CCS CONCEPTS • Information systems → World Wide Web; Web interfaces; Web log analysis.

show abstract

Section: Fashion Recommendationmentioning

confidence: 99%

Enhancing Fashion Recommendation with Visual Compatibility Relationship

Yin

et al. 2019

The World Wide Web Conference

View full text Add to dashboard Cite

show abstract

“…One of the efforts by item auxiliary information is to improve the accuracy of rating prediction. Some extra information has been explored to handle sparse data, for instance, audio features in music recommendations [11], [12], color feature extraction for online fashion shop [13], and document recommendation for online news recommendation [14], [15]. IMBD is an example of portal services that provides a product review about a movie coming from its customers.…”

Section: Introductionmentioning

confidence: 99%

“…CNN has been successfully applied in fashion ecommerce. Jaradat [13] proposes a novel approach based on the deep color classification to dig fashion model classification. He explores the relationship between color character and interest tendency in a product.…”

Section: Introductionmentioning

confidence: 99%

Convolutional-NN and Word Embedding for Making an Effective Product Recommendation Based on Enhanced Contextual Understanding of a Product Review

Hanafi¹,

Suryana

Basari

2019

Int. J. Adv. Sci. Eng. Inf. Technol.

View full text Add to dashboard Cite

E-commerce is one of the most popular service applications in the world in the last decade. It has become a revolutionary model from traditional shopping transaction to entire internet commerce. E-commerce needs essential artificial intelligence (AI) to provide the customer with information about a product, called a recommendation machine. Collaborative filtering is a model of a recommendation algorithm that relies on rating as the fundamental calculation to make a recommendation. It has been successfully implemented in e-commerce. Even so, this model has a weakness in sparse product data in which the rating number is very low or sparse. Mostly, only less than 3% of the total user population rate a product, leading to the rise of sparse data. A text sentence document is a part of customers' feedback that can be converted into a product rating. According to a traditional approach, bag of word and lexicon model are ignored in a contextual approach. This experiment, it developed a new model to increase the contextuality of text sentences, leading to a more effective rating prediction. We employed a kind of convolutional neural network to generate item latent factor vectors that could be incorporated with probabilistic matrix factorization to make rating prediction. Our method outperformed several previous works based on a metric evaluation using the Root Mean Squared Error (RMSE). In this experiment, we analyzed MovieLens and IMDB datasets, which contained a movie product review.

show abstract

“…1: An Instagram post from the fashion community. [7]. The text processing methods presented in this paper are meant to be integrated with computer vision models in the project.…”

Section: Introductionmentioning

confidence: 99%

Deep Text Mining of Instagram Data without Strong Supervision

Hammar

Jaradat

Dokoohaki

et al. 2018

2018 IEEE/WIC/ACM International Conference on Web Intelligence (WI)

Self Cite

View full text Add to dashboard Cite

With the advent of social media, our online feeds increasingly consist of short, informal, and unstructured text. This textual data can be analyzed for the purpose of improving user recommendations and detecting trends. Instagram is one of the largest social media platforms, containing both text and images. However, most of the prior research on text processing in social media is focused on analyzing Twitter data, and little attention has been paid to text mining of Instagram data. Moreover, many text mining methods rely on annotated training data, which in practice is both difficult and expensive to obtain. In this paper, we present methods for unsupervised mining of fashion attributes from Instagram text, which can enable a new kind of user recommendation in the fashion domain. In this context, we analyze a corpora of Instagram posts from the fashion domain, introduce a system for extracting fashion attributes from Instagram, and train a deep clothing classifier with weak supervision to classify Instagram posts based on the associated text.With our experiments, we confirm that word embeddings are a useful asset for information extraction. Experimental results show that information extraction using word embeddings outperforms a baseline that uses Levenshtein distance. The results also show the benefit of combining weak supervision signals using generative models instead of majority voting. Using weak supervision and generative modeling, an F1 score of 0.61 is achieved on the task of classifying the image contents of Instagram posts based solely on the associated text, which is on level with human performance. Finally, our empirical study provides one of the few available studies on Instagram text and shows that the text is noisy, that the text distribution exhibits the long-tail phenomenon, and that comment sections on Instagram are multi-lingual.

show abstract

Deep Cross-Domain Fashion Recommendation

Cited by 35 publications

References 14 publications

Enhancing Fashion Recommendation with Visual Compatibility Relationship

Enhancing Fashion Recommendation with Visual Compatibility Relationship

Convolutional-NN and Word Embedding for Making an Effective Product Recommendation Based on Enhanced Contextual Understanding of a Product Review

Deep Text Mining of Instagram Data without Strong Supervision

Contact Info

Product

Resources

About