As artificial intelligence technology matures, the use of generative AI chatbots (GAIC) has become widespread. However, this has been accompanied by the dilemma of problematic use. To ascertain the sociopsychological factors contributing to problematic use of generative AI chatbots (PUGAIC), this study conducted a survey targeting users who use generative AI chatbots (N = 303). The results indicate that social exclusion have indirect positive impacts on PUGAIC. Fear of negative evaluation, social anxiety, and loneliness play mediating role between them. It reflects that interactions with GAIC can complement impaired emotional needs in human-human interactions. In social context, privacy concerns do not function as a moderator, which is closely related to people’s attitude toward privacy and AI technology in China. Furthermore, this could present ethical dilemmas in the field of human–computer interaction. Accordingly, interventions to PUGAIC should help users understand the emotional nature of human–computer interactions and increase attention to privacy issues.

Self-supervised depth estimation has drawn much attention in recent years as it does not require labeled data but image sequences. Moreover, it can be conveniently used in various applications, such as autonomous driving, robotics, realistic navigation, and smart cities. However, extracting global contextual information from images and predicting a geometrically natural depth map remain challenging. In this paper, we present DLNet for pixel-wise depth estimation, which simultaneously extracts global and local features with the aid of our depth Linformer block. This block consists of the Linformer and innovative soft split multi-layer perceptron blocks. Moreover, a three-dimensional geometry smoothness loss is proposed to predict a geometrically natural depth map by imposing the second-order smoothness constraint on the predicted three-dimensional point clouds, thereby realizing improved performance as a byproduct. Finally, we explore the multi-scale prediction strategy and propose the maximum margin dual-scale prediction strategy for further performance improvement. In experiments on the KITTI and Make3D benchmarks, the proposed DLNet achieves performance competitive to those of the state-of-the-art methods, reducing time and space complexities by more than$62\%$and$56\%$at a resolution of$416 \times 128$, respectively. Extensive testing on various real-world situations further demonstrates the strong practicality and generalization capability of the proposed model.

Recovering the three-dimensional structure information from a monocular camera is significant for automated driving, robot navigation, and traffic safety assessment. Recent work has solved various tight issues on self-supervised monocular depth estimation of leveraging on-board videos, such as occlusion/disocclusion, dynamic objects, and scale inconsistent. Nevertheless, rare work focuses on the model’s prediction confidence and underlying relations between depths, while they are essential for a decision-making system and performance improvement, respectively. This paper proposes a novel scheme, that of correlation-aware structure, to dig into the relations between depths, converting the independent depths into a graph-like connected depth map. Subsequently, a Gaussian estimator is devised to predict the depth map and uncertainty map concurrently. The uncertainty map can show us problematic regions where it is difficult to predict, from which we further develop uncertainty-based strategies to improve the performance. Specifically, we propose a simple image preprocessing method to overcome the gradient locality issue caused by low-texture, especially in smooth roads and shadows. Also, to avoid the influence of high-uncertainty regions, we propose a solidity-aware mask to recognize the reliable pixels for training in the image. The experiments on the KITTI dataset show that our method results in a competitive performance in both depth and ego-motion estimation tasks compared with the state-of-the-art methods. Besides, additional experiments on the Make3D and Cityscapes datasets demonstrate our method’s strong generalization capability and practicality.

Scene perception is essential for driving decision-making and traffic safety. However, fog, as a kind of common weather, frequently appears in the real world, especially in mountain areas, making it difficult to accurately observe the surrounding environments. Therefore, precisely estimating the visibility under foggy weather can significantly benefit traffic management and safety. To address this, most current methods use professional instruments outfitted at fixed locations on the roads to perform the visibility measurement; these methods are expensive and less flexible. In this paper, we propose an innovative end-to-end convolutional neural network framework to estimate the visibility leveraging Koschmieder’s law and the image data. The proposed method estimates the visibility by integrating the physical model into the proposed framework, instead of directly predicting the visibility value via the convolutional neural network. Moreover, we estimate the visibility as a pixel-wise visibility map against those of previous visibility measurement methods which solely predict a single value for the entire image. Thus, the estimated result of our method is more informative, particularly in uneven fog scenarios, which can benefit to developing a more precise early warning system for foggy weather, thereby better protecting the intelligent transportation infrastructure systems and promoting their development. To validate the proposed framework, a virtual dataset, FACI, containing 3,000 foggy images in different concentrations, is collected using the AirSim platform, which is available athttps://github.com/coutyou/FoggyAirsimCityImages. Detailed experiments show that the proposed method achieves performance competitive to those of state-of-the-art methods.

Transcending the binary categorization of racist and xenophobic texts, this research takes cues from social science theories to develop a four-dimensional category for racism and xenophobia detection, namely stigmatization, offensiveness, blame, and exclusion. With the aid of deep learning techniques, this categorical detection enables insights into the nuances of emergent topics reflected in racist and xenophobic expression on Twitter. Moreover, a stage wise analysis is applied to capture the dynamic changes of the topics across the stages of early development of Covid-19 from a domestic epidemic to an international public health emergency, and later to a global pandemic. The main contributions of this research include, first the methodological advancement. By bridging the state-of-the-art computational methods with social science perspective, this research provides a meaningful approach for future research to gain insight into the underlying subtlety of racist and xenophobic discussion on digital platforms. Second, by enabling a more accurate comprehension and even prediction of public opinions and actions, this research paves the way for the enactment of effective intervention policies to combat racist crimes and social exclusion under Covid-19.

Online advertising and recommender systems often pose a multi-task problem, which tries to predict not only users’ click-through rate (CTR) but also the post-click conversion rate (CVR). Meanwhile, multi-functional information systems commonly provide multiple service scenarios for users, such as news feed, search engine and product suggestions. Users may leave similar interest information across various service scenarios. Thus the prediction/ranking model should be conducted in a multi-scene manner. This paper develops a unified r a nking m o del for this multi-task and multi-scene problem. Compared to previous works, our model explores independent/non-shared embeddings for each task and scene, which reduces the coupling between tasks and scenes. New tasks or scenes could be added easily. Besides, a simplified n e twork i s c h osen b e yond t h e embedding layer, which largely improves the ranking efficiency f o r online services. Extensive offline a n d o n line e x periments demonstrated the superiority of the proposed unified r a nking model.

Risky driving is a major cause of traffic accidents. In this paper, we propose a new method that recognizes risky driving behaviors purely based on vehicle speed time series. This method first retrieves the important distribution pattern of the sampled positive speed-change (value and duration) tuples for individual drivers within different speed ranges. Then, it identifies the risky drivers based on different patterns of drivers. Tests show the effectiveness of the proposed method. Since speed measurement is available on most of the newly build vehicles, this method can be easily implemented and used. The conclusion is useful to many traffic applications, e.g., driver training and insurance pricing.

Regression testing is essential for assuring the quality of a software product. Because rerunning all test cases in regression testing may be impractical under limited resources, test case prioritization is a feasible solution to optimize regression testing by reordering test cases for the current testing version. In this paper, we propose a new test case prioritization algorithm based on program changes and method (function) invocation relationship. Combining the estimated risk value of each program method (function) and the method (function) coverage information, the fault detection capability of each test case can be calculated. The algorithm reduces the prioritization problem to an integer linear programming (ILP) problem, and finally prioritizes test cases according to their fault detection capabilities. Experiments are conducted on 11 programs to validate the effectiveness of our proposed algorithm. Experimental results show that our approach is more effective than some well studied test case prioritization techniques in terms of average percentage of fault detected (APFD) values.

With the growing popularity of cloud storage service, how to ensure the integrity and availability of outsourced data has become a critical problem. To solve this, several remote data checking protocols are proposed to prove the data possession and retrievability. However, these methods either incur large overhead or are unable to repair the remote corruptions. In this paper, we propose the multi-replicas based data integrity verification and recovery (MRVR) scheme to realize public auditing and rapid recovery in multi-cloud environment. The protocols of replication-based verification and recovery are proposed based on the techniques of homomorphism, bilinear map and the binary aggregation tree (BAT). Through security analysis, MRVR is proved to be secure and privacy preserving with high data recoverability and availability. Finally, we analyze the system performance from both computation and communication overheads, and compare the algorithm efficiencies in the courses of verification and recovery with relevant schemes.

One primary challenge of applying access control methods in cloud computing is to ensure data security while supporting access efficiency, particularly when adopting multiple access control policies.Many existing works attempt to propose suitable frameworks and schemes to solve the problems, however, these proposals only satisfy specified use cases.In this paper, we take XACML as the policy language and build up a logical model.Based on this, we introduce the fine-grained data fragment algorithm to optimize the policies, whose resource property represents physical meaningful data blocks.Data are organized in a tree structure, where each leaf node represents a minimal physical meaningful data block, and internal nodes are combined data types.This method can eliminate conflicts and redundancies among rules and policies, thus to refine the policy set and achieve fine-grained access control.Our approach can also be applied to processing multi-types of data, and experiments are carried out to show the improvements of efficiencies.

One primary challenge of enforcing access control in cloud computing is how to ensure access with high efficiency while preserving data security. This paper proposes a fine-grained access control method for cloud resources. The basic idea is to use XACML as access control language and to optimize policies by data fragmentation and policy refinement algorithms. Through data fragmentation, the accessible resources are divided into disjoint data blocks, and each of them will be combined with a set of policy rules. This helps to refine the policy and to avoid data leakage caused by rule conflicting on the resource intersections. Finally, the disjoint data blocks and the optimized policy are distributed in the three-layered cloud, and the decision to a request is made by rule matching on a specific resource rather than traversing the whole policy rules. Experiments show that our proposal enjoys higher efficiency in cloud-based access control.