Within the broader scope of applications of artificial intelligence to financial forecasting, this paper studies how to deal with nonlinearity and high noise for stock price prediction. We apply BPNN to investigate appropriate configuration settings for long-term stock prediction. Through systematically adjusting hyperparameters like the data span, the hidden-layer structure, or the number of neurons, we show that the most appropriate model for capturing Nestlé stock’s long-run trend characteristics is the one trained on 10 years of data, with 16 neurons over two hidden layers. This setting is not only an inherent regularity of a mature company’s behavior, but it also has robustness as we have done many comparisons. It makes it part of using neural networks in an attempt to achieve financial forecasts.

Show more

Recent developments in recommender systems have increasingly employed deep learning methodologies to confront long-standing challenges, including the modeling of intricate user–item interactions, the incorporation of temporal dynamics, and the mitigation of exposure bias. This study reviews and extends insights from four representative approaches. First, the Convolutional Transformer Neural Collaborative Filtering (CTNCF) model combines convolutional neural networks with Transformer architectures to capture both localized and long-range dependencies within user–item representations, thereby surpassing the performance of conventional Neural Collaborative Filtering (NCF). Second, the Neural Tensor Factorization (NTF) framework advances classical tensor factorization by embedding recurrent and multilayer neural components, enabling the representation of time-varying preferences and nonlinear interactions among latent factors. Third, the Deep Interest Network (DIN) introduces a local activation mechanism that adaptively models user interests in click-through rate prediction, effectively overcoming the limitations of fixed-length embeddings in capturing heterogeneous behavioral patterns; notably, this model has been deployed at scale in industrial advertising contexts. Finally, recent work addressing de-exposure bias in NCF incorporates reward signals derived from the LinUCB algorithm into the neural recommendation process, thereby enhancing both fairness and predictive accuracy by increasing the visibility of underexposed items. Taken together, these contributions illustrate the progression of neural recommender systems from static factorization paradigms toward dynamic, adaptive, and fairness-oriented frameworks, offering both theoretical contributions and practical value for the design of large-scale recommendation platforms.

Show more

In the context of the rapid development of electrification and intelligentization in the global automotive industry, vehicle structure design is undergoing a paradigm shift. Modern automotive structures face the increasing pressure of safety requirements and the dual pressures of improving energy efficiency and reducing emissions while extending driving range. This paper systematically analyzes the main characteristics of modern automotive structures and focuses on the technical route and key challenges of lightweighting and safety integrated optimization. The evolution of methodologies and the essential value of vehicle structural optimization are presented. Current optimization strategies mainly involve topology optimization, dimensional and shape optimization, and material selection optimization. The relevant technologies for optimization include lightweight design, crashworthiness enhancement, and manufacturing constraint consideration optimization. Main challenges include high computational cost and efficiency, multi-objective trade-offs, few experimental verifications, and difficulties in standardization and industrialization. This paper emphasizes the interdisciplinary collaboration between different disciplines.

Show more

The rapid expansion of the Internet of Things (IoT) brings an abrupt breakthrough in intelligent homes, and allows people to live in a smarter and easier way. In this paper, some basic technologies of intelligent home based on IoT are explored, such as device communication protocols, data processing flow and smart control algorithms. Although these technologies have great potential applications in an intelligent home, there are some significant problems that need to be solved in real life. The different compatibility between devices is the biggest problem that exists now. Meanwhile, data security and protection for users’ privacy remain an important problem as well. The system energy consumption and reliability are another barrier that should be improved to satisfy the diversified interests at home. The conclusion can be made that in the future, the research topics should include standardizing techniques and specifications, improving the system security and developing machine learn ability of intelligent devices which can be matched to the intelligent home requirements. It is believed these improvements are necessary for the efficient development and wider application of smart home systems.

Show more