ABSTRACT
The use of Artificial Intelligence (AI) techniques has become prevalent with almost all e-commerce sites. Following the trend, today's Fashion recommender systems are no exception, which widely use AI for catering to multiple use-cases. [3][4]. However this is still quite far from replicating the personalized shopping experience in a fashion store. Through this paper and demonstration we try and replicate the offline experience in an online world by using a Natural Language chatbot powered by multiple deep learning models. Athena, the personal shopping adviser in the presented system enables user interaction in natural language via a web-browser based application (extensible to a mobile application). The application is powered by two underlying machine learning components - Recommendence and Fashionsence. The recommendation system Recommendence is an ensemble of deep learning based and graph based collaborative filtering recommenders and is able to provide personalized fashion recommendations to the users. The Fashionsence component is a deep learning based model which can learn a fashion sense from online, trend and stylist data to recommend a trending fashionable set of items from the product inventory. This entire system is integrated via an orchestrating layer.
- {n. d.}. tf.losses.mean_pairwise_squared_error | TensorFlow. https://www.tensorflow.org/api_docs/python/tf/losses/mean_pairwise_squared_error. (Accessed on 09/29/2018).Google Scholar
- Gediminas Adomavicius, Bamshad Mobasher, Francesco Ricci, and Alexander Tuzhilin. 2011. Context-Aware Recommender Systems. AI Magazine 32, 3 (oct 2011), 67.Google ScholarCross Ref
- Chester Gray, Meghan Beattie, Helena Belay, Sarah Hill, and Nicolette Lerch. 2015. Personalized online search for fashion products. In 2015 Systems and Information Engineering Design Symposium. IEEE.Google ScholarCross Ref
- Xintong Han, Zuxuan Wu, Yu-Gang Jiang, and Larry S Davis. 2017. Learning fashion compatibility with bidirectional lstms. In Proceedings of the 2017 ACM on Multimedia Conference. ACM, 1078--1086. Google ScholarDigital Library
- Malcolm Slaney and Michael Casey. 2008. Locality-sensitive hashing for finding nearest neighbors {lecture notes}. IEEE Signal processing magazine 25, 2 (2008), 128--131.Google ScholarCross Ref
- Christian Szegedy, Wei Liu, Yangqing Jia, Pierre Sermanet, Scott Reed, Dragomir Anguelov, Dumitru Erhan, Vincent Vanhoucke, and Andrew Rabinovich. 2015. Going deeper with convolutions. In Proceedings of the IEEE conference on computer vision and pattern recognition. 1--9.Google ScholarCross Ref
Index Terms
- Recommendence and Fashionsence: Online Fashion Advisor for Offline Experience
Recommendations
Using Implicit Feedback for Neighbors Selection: Alleviating the Sparsity Problem in Collaborative Recommendation Systems
WebMedia '17: Proceedings of the 23rd Brazillian Symposium on Multimedia and the WebThe most popular Recommender systems (RSs) employ Collaborative Filtering (CF) algorithms where users explicitly rate items. Based on these ratings, a user-item rating matrix is generated and used to select the items to be recommended for a target user. ...
A General Rating Recommended Weight-Aware Model for Recommendation System
HCC 2016: Revised Selected Papers of the Second International Conference on Human Centered Computing - Volume 9567In recommendation system, the ratings represent the users' preference and play an important role in recommending items to users. However, the ratings of items may be influenced by many factors, such as time the latest ratings are more able to reflect ...
Exploiting Social Tagging in a Web 2.0 Recommender System
Recommender systems help users cope with information overload by using their preferences to recommend items. To date, most recommenders have employed users' ratings, information about the user's profile, or metadata describing the items. To take ...
Comments