Explore the vast range of titles available.

The meaning of privacy has been much disputed throughout its history in response to wave after wave of new technological capabilities and social configurations. The current round of disputes over privacy fuelled by data science has been a cause of despair for many commentators and a death knell for privacy itself for others. We argue that privacy's disputes are neither an accidental feature of the concept nor a lamentable condition of its applicability. Privacy is essentially contested. Because it is, privacy is transformable according to changing technological and social conditions. To make productive use of privacy's essential contestability, we argue for a new approach to privacy research and practical design, focused on the development of conceptual analytics that facilitate dissecting privacy's multiple uses across multiple contexts. This article is part of the themed issue ‘The ethical impact of data science’.

In the era of \"Internet plus,\" the world economy is becoming more and more globalized and informationalized. China's enterprises are facing unprecedented opportunities for their operation and development. However, it is also facing the financial uncertainties brought about by the fluctuations of the general economic environment, and the company is facing increasing financial risks. The reason why most enterprises encounter a serious financial crisis or even close down in the later stage is that they do not pay full attention to the initial financial problems and do not take effective measures to deal with the crisis in time. Financial risk warning has become an important part of modern enterprise financial management. This paper mainly puts forward the optimized BP neural system as the financial early warning model and ensures its high prediction accuracy. In the research, the operation principle and related reasoning process of the model are described, its shortcomings are analyzed, and solutions are put forward. Through the financial risk analysis of listed companies from 2017 to 2020, we find that the correct rate of the prediction results of the financial distress of normal companies in the selected companies based on the optimized BPNN has reached more than 80%, which proves the effectiveness of the optimized BPNN.

In the competitive retail industry of the digital era, data-driven insights into gender-specific customer behavior are essential. They support the optimization of store performance, layout design, product placement, and targeted marketing. However, existing computer vision solutions often rely on facial recognition to gather such insights, raising significant privacy and ethical concerns. To address these issues, this paper presents a privacy-preserving customer analytics system through two key strategies. First, we deploy a deep learning framework using YOLOv9s, trained on the RCA-TVGender dataset. Cameras are positioned perpendicular to observation areas to reduce facial visibility while maintaining accurate gender classification. Second, we apply AES-128 encryption to customer position data, ensuring secure access and regulatory compliance. Our system achieved overall performance, with 81.5% mAP@50, 77.7% precision, and 75.7% recall. Moreover, a 90-min observational study confirmed the system’s ability to generate privacy-protected heatmaps revealing distinct behavioral patterns between male and female customers. For instance, women spent more time in certain areas and showed interest in different products. These results confirm the system’s effectiveness in enabling personalized layout and marketing strategies without compromising privacy.

To guarantee the safe operation of machinery and reduce its maintenance costs, estimating its remaining useful life (RUL) is a crucial task. Hence, in this study, a multi-scale memory-enhanced prediction method is proposed to describe fully characteristics of the data. This method is based on a deep learning algorithm and is designed to estimate the RUL of aircraft engines. To handle the complex and multi-fault operating conditions with uncertain properties in RUL estimation, a hybrid model that combines a multi-scale deep convolutional neural network and long short-term memory is presented. Experimental verification was carried out with the Commercial Modular Aero-Propulsion System Simulation dataset from NASA. Compared with multi-scale deep convolutional and long short-term memory networks, the hybrid model performed more efficiently. Furthermore, compared with other state-of-the-art methods, the multi-scale memory-enhanced prediction method can achieve better prognostics, especially for equipment with multiple operating conditions and failure modes.

As an integral part of intelligent transportation system, vehicle detection and tracking system is of great research significance and practical application value. In this paper, based on the mixed Gaussian background model, the detection target is segmented by the different methods, and the most matching target track is found by using the location information and color information of the detection target, so as to realize the vehicle tracking. The experiment results show that for the same target, the centroid distance is less than 0.2, the color distance of HSV (hue saturation value) is less than 0.3, the centroid distance of different targets is less than 0.2, the HSV distance is less than 0.3, and the rest are distributed to some extent. When the centroid distance is 0.01, 0.02, 0.03, 0.04, 0.05 and 0.06, respectively, the matching results are 250, 150, 100, 50, 25 and 10, respectively; when the HSV color distance is 0.02, 0.06, 0.1, 0.14 and 0.18, respectively, the matching results are 160, 200, 100, 80 and 50, respectively. Therefore, for the normalized distance between the same targets, including the centroid distance and HSV color, in each possible matching area, the greater the distance is, the less the distribution of matching results. Experimental verification shows that when the vehicle is detected in the detection area, the effective contour is sequentially accessed and tracked through the memory pointer, and the relatively accurate contour of the moving vehicle can be obtained through the improved Gaussian mixture model. The vehicle detection algorithm based on regional method has high real-time accuracy and strong practical value, can meet the needs of intelligent transportation system, and has strong practical value.

Image classification can effectively manage and organize images, laying a good foundation for the work in multiple fields of image processing. With the rise of Internet technology and social networks, the number of digital images has increased dramatically and there are more and more applications. People also use intuitive pictures instead of words when expressing emotions and information. A large number of digital images need to be managed, analyzed, and retrieved. This urgently requires more efficient and accurate image classification technology. Deep learning is a learning method that extends traditional neural networks, simulating the process of gradual abstraction of human brain cognition. The number of hidden layers is deepened, and features can be learned automatically. This paper has completed the traditional image noise detection and segmentation experiment based on convolutional neural network. We introduced the various parts of the convolutional neural network, including the convolutional layer, activation function, fully connected layer, etc. Noise detection is completed based on Fast RCNN on the MSTAR data set containing the background. The model fully combines the advantages of the isotropic model, the Perona-Malik model, and the second-order directional derivative. In order to better maintain the edge structure information of the image, the structure information is extracted based on the original noise image, and the improved ID model and the improved PM model are directionally diffused along the edge tangent direction of the original noise image. Considering the local structure information of the image, we use the slice similarity modulus value as the edge detector of the improved PM model, and use the slice similarity modulus value to construct a new weighting function to adaptively balance the relative weights of the improved ID model and the improved PM model. Simulation and measured data verify the effectiveness of this network in removing image coherent speckle noise. We compare and analyze it with existing denoising methods. The use of visual evaluation and objective evaluation indicators to evaluate the denoising effect and calculation efficiency shows the advantages of the network in this paper in terms of denoising effect, calculation time and space complexity.

The investigation of how to accurately predict the sale price of houses is the main objective of our work. Accurate secondhand housing price appraisal is critical in secondhand housing deals, mortgages, and risk assessment. Due to the complex composition of real estate prices, the difficulty of obtaining data and the lack of effective algorithms, the accurate appraisal of housing prices is still a challenge. Based on the hedonic model, the anchoring effect is added to the structure and location characteristics in this work. The 2SFCA algorithm is introduced into the location feature index to filter the influence of the accessibility index. Our model was trained using a variety of machine learning models, such as linear regression and random forest, and the results were evaluated to determine a suitable algorithm for building a secondhand housing transaction price forecasting model. The results showed that the prediction accuracy of the price prediction model could be improved by adding the facility accessibility index, and when the anchoring effect is added to the price prediction model, the prediction accuracy of the model could increase to 0.89. In comparing the results of various machine learning algorithms, we found that the ETR, RFR, and GBR models had better prediction results, and the accuracy rate could reach 0.9. In the end, a case study in Shenzhen was utilized to show that our proposed framework for predicting the price of secondhand houses, which integrated behavioral economics, hedonic price theory, and machine learning algorithms, was practical and efficient and can effectively improve the efficiency and accuracy of the evaluation.

The research on state estimation for quadruped robots is critical. Its result passed to motion controller makes the robot navigate autonomously and adjust the gait to a more stable motion. The current research depends on a multi-sensor fusion of cameras, lidars or other proprioceptive sensors, such as Inertial Measurement Unit (IMU) and encoders. The high-frequency data are generally derived from body sensors, which is to be fused with data from external sensors directly, or preprocessed with EKF first. Due to its unguaranteed convergence and robustness of tracking state mutations, EKF is insufficient. Therefore, we study state estimation for hydraulic quadruped robot based on the fusion of IMU measurement and leg odometry in this paper, and Invariant Extended Kalman Filter (IEKF) is successfully applied to quadruped robots by using this method. Besides, neural networks are utilized to train the weight functions of foot force and the state of leg odometry, and our trained functions improve the accuracy of observation compared with common weight average methods. Finally, our experiments of accuracy show that the root mean square error of our method is significantly reduced and the absolute trajectory error is reduced by 30% compared to traditional IEKF. The algorithm achieves the drift per distance travelled below 4 cm/m. Moreover, it has good robustness.

Crowd counting (CC) and density estimation are crucial for ensuring public safety and security in surveillance videos with large audiences. As computer vision-based scene interpretation advances, automatic analysis of crowd situations is becoming increasingly prevalent. However, existing crowd analysis algorithms may not accurately interpret the video footage. To address this challenge, we propose a new approach called SMOHDL-CCA. This approach combines a Slime Mold Optimization algorithm with a Hybrid Deep Learning Enabled CC Approach. Our system uses the SMO algorithm with an optimized neural network search network (NASNet) model as the front-end to take advantage of transfer learning and flexible characteristics. The back-end model employs Dilated Convolutional Neural Networks, and the hyperparameter tuning process is done using the Chicken Swarm Optimization algorithm. Given a crowded video input frame, our SMOHDL-CCA model estimates the density map of the image. Each pixel value indicates the crowd density at the corresponding location in the picture. The final crowd count is obtained by summing all the values in the density map. We evaluated our proposed approach using three standard datasets. Furthermore, the state-of-the-art combining the proposed SMOHDL-CCA model achieves comparable performance such as improved precision is 96.97%, recall is 96.94%, and F1 score is 96.61%, reduced mean squared error of 61.15 values for the NWPU-crowd, UCF_QNRF, and World Expo datasets.

Computer vision methods for non-contact detection of herd pigs could help detect early disease and reduce mortality rates by analyzing pig behavior. Due to the limitation of breeding space and cost, the unit breeding area is relatively dense, making it difficult to detect all pigs for a long time without interruption. In order to improve the detection performance, this paper proposes an end-to-end efficient and accurate herd pig detection framework based on YOLOv7 target detection model, which is named Ghost-YOLOv7-SIoU. In this framework, the feature extraction backbone network consists of a series of directly connected efficient layer aggregation networks (ELAN) and downsampling modules. The neck network contains a feature pyramid network and path aggregation network. Ghost convolution is adopted to replace the 3 × 3 standard convolution of the ELAN module in backbone network and the scaled-up ELAN module in neck network to obtain rich features while reducing the parameter number and computational effort. Furthermore, to speed up the model convergence and improve the model robustness and accuracy, SIoU loss is used for bounding box regression in the training stage. On the VOC2012 dataset, the number of parameters and floating-point operations decreased by 13.4% and 15.7% compared to YOLOv7, with comparable detection accuracy. Additionally, the number of parameters and floating-point operations decreased by 13.7% and 16.1% on our pig dataset. Ghost-YOLOv7-SIoU is superior to YOLOV4-CSP and YOLOR-CSP in accuracy. Experimental results demonstrate the effectiveness of the proposed method in improving the efficiency of model detection while ensuring detection accuracy.