No abstract

The goal of personalized history-based recommendation is to automatically output a distribution over all the items given a sequence of previous purchases of a user. In this work, we present a novel approach that uses a recurrent network for summarizing the history of purchases, continuous vectors representing items for scalability, and a novel attention-based recurrent mixture density network, which outputs each component in a mixture sequentially, for modelling a multi-modal conditional distribution. We evaluate the proposed approach on two publicly available datasets, MovieLens-20M and Rec-Sys15. The experiments show that the proposed approach, which explicitly models the multi-modal nature of the predictive distribution, is able to improve the performance over various baselines in terms of precision, recall and nDCG.

Position bias, the phenomenon whereby users tend to focus on higher-ranked items of the search result list regardless of the actual relevance to queries, is prevailing in many ranking systems. Position bias in training data biases the ranking model, leading to increasingly unfair item rankings, click-through-rate (CTR), and conversion rate (CVR) predictions. To jointly mitigate position bias in both item CTR and CVR prediction, we propose two positionbias-free CTR and CVR prediction models: Position-Aware Click-Conversion (PACC) and PACC via Position Embedding (PACC-PE). PACC is built upon probability decomposition and models position information as a probability. PACC-PE utilizes neural networks to model product-specific position information as embedding. Experiments on the E-commerce sponsored product search dataset show that our proposed models have better ranking effectiveness and can greatly alleviate position bias in both CTR and CVR prediction. CCS CONCEPTS• Applied computing → Electronic commerce; • Computing methodologies → Machine learning.

The lack of high-quality labeled training data has been one of the critical challenges facing many industrial machine learning tasks. To tackle this challenge, in this paper, we propose a semi-supervised learning method to utilize unlabeled data and user feedback signals to improve the performance of ML models. The method employs a primary model M ain and an auxiliary evaluation model Eval, where M ain and Eval models are trained iteratively by automatically generating labeled data from unlabeled data and/or users feedback signals. The proposed approach is applied to different text classification tasks. We report results on both the publicly available Yahoo! Answers dataset and our e-commerce product classification dataset. The experimental results show that the proposed method reduces the classification error rate by 4% and up to 15% across various experimental setups and datasets. A detailed comparison with other semi-supervised learning approaches is also presented later in the paper. The results from various text classification tasks demonstrate that our method outperforms those developed in previous related studies.

With development of smart grid, the stable operation of grid has put forward higher requirements for system dispatch. In particular, short-term load forecasting of power systems is a key factor of power grid management systems, which is related to the safety, economy, and stable operation of the smart grid. However, short-term electricity forecasting is affected by many external factors. It has complex characteristics, especially non-linear relationships, so it cannot be accurately predicted. Recently, Recurrent Neural Network based models have good performance in electricity forecasting because of their excellent ability to capture non-linear relationships. However, they cannot fully capture historical information, especially local historical information, which has an impact on prediction accuracy. In order to address these problems, we propose a scheme by combining STL decomposition and GRU model. Specifically, we first decompose the original time series into three different components by STL. The decomposition results are separately imported into the main prediction module, which uses two GRU networks with different structures to obtain the local and global dependencies of the data. We also add an autoregressive method to make the model more robust. The proposed scheme is validated based on real-world data, and the simulation results show that our proposed method can perfectly capture local and global information and achieve higher prediction accuracy than traditional models.