Title: HRM: Learning Hierarchical Representation Model for Next Basket Recommendation.
Authors: Pengfei Wang
Abstract: Next basket recommendation is a crucial task in market bas- ket analysis. Given a user’s purchase history, usually a sequence of transaction data, one attempts to build a recom- mender that can predict the next few items that the us- er most probably would like. Ideally, a good recommender should be able to explore the sequential behavior (i.e., buy- ing one item leads to buying another next), as well as ac- count for users’ general taste (i.e., what items a user is typically interested in) for recommendation. Moreover, these two factors may interact with each other to influence users’ next purchase. To tackle the above problems, in this pa- per, we introduce a novel recommendation approach, name- ly hierarchical representation model (HRM). HRM can well capture both sequential behavior and users’ general taste by involving transaction and user representations in prediction. Meanwhile, the flexibility of applying different aggregation operations, especially nonlinear operations, on representations allows us to model complicated interactions among different factors. Theoretically, we show that our model subsumes several existing methods when choosing proper aggregation operations. Empirically, we demonstrate that our model can consistently outperform the state-of-the-art baselines under different evaluation metrics on real-world transaction data.
Running with RecBole¶
embedding_size (int): The embedding size of users and items. Defaults to
high_order (int): The last N items . Defaults to
pooling_type_layer_1 (str): The type of pooling in the first floor include average pooling and max pooling . Defaults to
pooling_type_layer_2 (str): The type of pooling in the second floor include average pooling and max pooling . Defaults to
dropout_prob (float): The dropout rate. Defaults to
loss_type (str): The type of loss function. If it set to
'CE', the training task is regarded as a multi-classification task and the target item is the ground truth. In this way, negative sampling is not needed. If it set to
'BPR', the training task will be optimized in the pair-wise way, which maximize the difference between positive item and negative item. In this way, negative sampling is necessary, such as setting
training_neg_sample_num = 1. Defaults to
'CE'. Range in
A Running Example:
Write the following code to a python file, such as run.py
from recbole.quick_start import run_recbole run_recbole(model='HRM', dataset='ml-100k')
reproducibility=False, the training speed of HRM can be greatly accelerated.
Tuning Hyper Parameters¶
If you want to use
HyperTuning to tune hyper parameters of this model, you can copy the following settings and name it as
learning_rate choice [0.001] embedding_size choice  high_order choice [1,2,4] dropout_prob choice [0.2] pooling_type_layer_1 choice ["max","average"] pooling_type_layer_2 choice ["max","average"]
Note that we just provide these hyper parameter ranges for reference only, and we can not guarantee that they are the optimal range of this model.
Then, with the source code of RecBole (you can download it from GitHub), you can run the
run_hyper.py to tuning:
python run_hyper.py --model=[model_name] --dataset=[dataset_name] --config_files=[config_files_path] --params_file=hyper.test
For more details about Parameter Tuning, refer to Parameter Tuning.
If you want to change parameters, dataset or evaluation settings, take a look at