Deep Matrix Factorization Models for Recommender Systems

Authors: Hong-Jian Xue, Xinyu Dai, Jianbing Zhang, Shujian Huang, Jiajun Chen

IJCAI 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental The experimental results show the effectiveness of both our proposed model and the loss function. On several benchmark datasets, our model outperformed other state-of-the-art methods. We also conduct extensive experiments to evaluate the performance within different experimental settings.
Researcher Affiliation Academia Hong-Jian Xue, Xin-Yu Dai, Jianbing Zhang, Shujian Huang, Jiajun Chen National Key Laboratory for Novel Software Technology; Nanjing University, Nanjing 210023, China Collaborative Innovation Center of Novel Software Technology and Industrialization, Nanjing 210023, China xuehj@nlp.nju.edu.cn, {daixinyu,zjb,huangsj,chenjj}@nju.edu.cn
Pseudocode Yes Algorithm 1 DMF Training Algorithm With Normalized Cross Entropy
Open Source Code No We implemented our proposed methods based on Tensorflow3, which will be released publicly upon acceptance.
Open Datasets Yes We evaluate our models on four widely used datasets in recommender systems: Movie Lens 100K(ML100k), Movie Lens 10M(ML1m), Amazon music(Amusic), Amazon movies(Amovie). They are publicly accessible on the websites 1 2. 1https://grouplens.org/datasets/movielens/ 2http://jmcauley.ucsd.edu/data/amazon/
Dataset Splits Yes We held-out the latest interaction as a test item for every user and utilize the remaining dataset for training. ... To determine hyper-parameters of DMF methods, we randomly sampled one interaction for each user as the validation data and tuned hyper-parameters on it.
Hardware Specification No The paper does not provide specific hardware details (e.g., CPU/GPU models, memory, or cloud instance types) used for running the experiments.
Software Dependencies No The paper mentions 'TensorFlow3' but does not specify a version number for TensorFlow or any other software dependencies.
Experiment Setup Yes When training our models, we sampled seven negative instances per positive instance. For neural network, we randomly initialized model parameters with a Gaussian distribution (with a mean of 0 and standard deviation of 0.01), optimizing the model with mini-batch Adam [Kingma and Ba, 2014]. We set the batch size to 256, and set the learning rate to 0.0001.