1. Billion-scale Commodity Embedding for E-commerce Recommendation in Alibaba
Jizhe Wang, Pipei Huang, Huan Zhao, Zhibo Zhang, Binqiang Zhao, Dik Lun Lee
The billion-scale data inTaobaocreates three major challenges to Taobao's RS:scalability,sparsityandcold start. In this paper, we present our technical solutions to address these three challenges. The methods are based on thegraph embeddingframework. We first construct an item graph from users' behavior history. Each item is then represented as a vector using graph embedding. The item embeddings are employed to compute pairwise similarities between all items, which are then used in the recommendation process. We propose two aggregation methods to integrate the embeddings of items and the corresponding side information. Experimental results from offline experiments show that methods incorporating side information are superior to those that do not. Using online A/B test, we show that the online Click-Through-Rate (CTRs) are improved comparing to the previous recommendation methods widely used in Taobao.
2. Cross-Domain Recommendation for Cold-Start Users via Neighborhood Based Feature Mapping
Xinghua Wang, Zhaohui Peng, Senzhang Wang, Philip S. Yu, Wenjing Fu, Xiaoguang Hong
Although Cross-Domain Collaborative Filtering (CDCF) is proposed for effectively transferring users' rating preference across different domains, it is still difficult for existing CDCF models to tackle the cold-start users in the target domain due to the extreme data sparsity. In this paper, we propose a Cross-Domain Latent Feature Mapping (CDLFM) model for cold-start users in the target domain. Firstly, we take the users' similarity relationship on rating behaviors into consideration and propose the Matrix Factorization by incorporating User Similarities (MFUS). Next, we propose a neighborhood based gradient boosting trees method to learn the cross-domain user latent feature mapping function. For each cold-start user, we learn his/her feature mapping function based on the latent feature pairs of those linked users who have similar rating behaviors with the cold-start user in the auxiliary domain.
3. Word2Bits - Quantized Word Vectors
Word vectors require significant amounts of memory and storage, posing issues to resource limited devices like mobile phones and GPUs. We show that high quality quantized word vectors using 1-2 bits per parameter can be learned by introducing a quantization function into Word2Vec. We furthermore show that training with the quantization function acts as a regularizer. We train word vectors on English Wikipedia (2017) and evaluate them on standard word similarity and analogy tasks and on question answering (SQuAD). Our quantized word vectors not only take 8-16x less space than full precision (32 bit) word vectors but also outperform them on word similarity tasks and question answering.
4. The Impact of Semantic Context Cues on the User Acceptance of Tag Recommendations: An Online Study
Dominik Kowald, Paul Seitlinger, Tobias Ley, Elisabeth Lex
In this paper, we present the results of an online study with the aim to shed light on the impact that semantic context cues have on the user acceptance of tag recommendations. We validated and verified the hypothesis that semantic context cues have a higher impact on the user acceptance of tag recommendations in a collaborative tagging setting than in an individual tagging setting. With this paper, we contribute to the sparse line of research presenting online recommendation studies.
5. Investigating the utility of the weather context for point of interest recommendations
Christoph Trattner, Alexander Oberegger, Leandro Marinho, Denis Parra
In this paper, we contribute to this area of research by presenting the novel results of a study that aims to recommend POIs based on weather data. To this end, we have expanded the state-of-the-art Rank-GeoFM POI recommender algorithm to include additional weather-related features such as temperature, cloud cover, humidity and precipitation intensity. We show that using weather data not only significantly improves the recommendation accuracy in comparison to the original method, but also outperforms its time-based variant. Furthermore, we investigate the magnitude of the impact of each feature on the recommendation quality. Our research clearly shows the need to study weather context in more detail in light of POI recommendation systems.
6. A collection of resources for Recommender Systems (RecSys)