No abstract

No abstract

Click-through Rate (CTR) prediction has become one of the core tasks of the recommendation system and its online advertising with the development of e-commerce. In the CTR prediction field, different features extraction schemes are used to mine the user click behavior to achieve the maximum CTR, which helps the advertisers maximize their profits. At present, achievements have been made in CTR prediction based on Deep Neural Network (DNN), but insufficiently, DNN can only learn high-order features combination. In this paper, Product & Cross supported Stacking Network with LightGBM (PCSNL) is proposed for CTR prediction to solve such problems. Firstly, the L 1 and L 2 regularizations are imposed on Light Gradient Boosting Machine (LightGBM) to prevent overfitting. Secondly, the method of vector-wise feature interactions is added to product layer in product network to learn second-order feature combinations. Lastly, feature information is fully learned through the cross network, product network and stacking network in PCSNL. The online ads CTR prediction datasets released by Huawei and Avazu on the Kaggle platform are involved for experiments. It is shown that the PCSN model and PCSNL have better performance than the traditional CTR prediction models and deep learning models.

Denoisers trained with synthetic noises often fail to cope with the diversity of real noises, giving way to methods that can adapt to unknown noise without noise modeling or ground truth. Previous image-based method leads to noise overfitting if directly applied to temporal denoising, and has inadequate temporal information management especially in terms of occlusion and lighting variation. In this paper, we propose a general framework for temporal denoising that successfully addresses these challenges. A novel twin sampler assembles training data by decoupling inputs from targets without altering semantics, which not only solves the noise overfitting problem, but also generates better occlusion masks by checking optical flow consistency. Lighting variation is quantified based on the local similarity of aligned frames. Our method consistently outperforms the prior art by 0.6-3.2dB PSNR on multiple noises, datasets and network architectures. Stateof-the-art results on reducing model-blind video noises are achieved.

The traditional model of grey nearness degree of incidence contains some inherent limitations in the calculation of data sequences. It does not consider the impacts of certain data on degree of incidence when there are significant differences in orders of magnitude between adjacent data in the same sequence, and big errors may occur in the calculation of some special oscillation sequences. In response to these problems, we propose a new improved method, which uses the characteristics of the model of grey nearness degree of incidence and introduces a neural network algorithm to define a grey neural network-nearness degree of incidence. Thereby, a model of nearness degree of incidence is established based on grey neural network. Then we apply a new model to the field of data mining. According to the clustering algorithm, we take all the degrees of incidence as the variables of the distance metric function, and use the clustering algorithm of data mining for data analysis. Finally, through simulation experiments, we verify the effectiveness of the clustering algorithm under the new distance metric definition. The experimental results show that, compared with other methods, the computational outcomes of the improved model are more consistent with the actual situation. The cluster algorithm with the model used can deliver results that have a high accuracy, so the new model can be applicated in a wide range of fields.INDEX TERMS Grey neural network, nearness degree of incidence, data mining, clustering algorithm.

Based on the theories of fuzzy set and fuzzy conversion, the method of fuzzy comprehensive appraisal is a decision-making process which combines qualitative analysis and quantitative analysis and can be used to forecast risk of electric power engineering projects. Using the method of AHP to establish risk early-warning indicators system and method of fuzzy comprehensive appraisal to establish risk early-warning model, the paper constitutes risk early-warning system of electric power engineering projects. A case from western China is applied to prove the validity of the risk early-warning system.

An effective multi-feature extraction method for well-log curve line plotted manually. Several log curve line features are used to extract the well log curve information in the proposed method: gray value similarity feature, connected region Length feature, projection feature and distance and angular feature. The experimental results show the good effect of the method for the well log curve information extraction.

No abstract