教学课程资源库 更多>>

关键词： Malware   Encryption   Monitoring   Electronic mail   Behavioral sciences   Computer science   ADS (Alternate Data Streams)   ransomware   file encryption   file encryption  

摘要： Ransomware attacks have become widespread in the last few years and have affected many critical industries and infrastructures. Unfortunately, there are no recovery tools that can effectively defend against all types of ransomware. Approaches, such as frequent data backups, have several drawbacks. They are expensive in terms of resources and trained technical staff. Therefore, it is much more challenging and cost-consuming for average users and small business owners to survive ransomware attacks. To provide an easy-to-use tool for a broader population of users and businesses, we propose a novel ransomware defense mechanism that can be conveniently deployed in modern Windows systems which have over 76% market share as of 2022. The uniqueness of our approach is to fight malware like malware. We leverage Alternate Data Streams, which are sometimes used by malicious applications, to design and implement a data protection method that misleads the ransomware into attacking only file "shells" instead of the actual file content. We have evaluated our approach against different cryptographic ransomware. The results show that our approach is usable, efficient, and effective.

摘要： The Chinese heroin generation is a unique yet less discussed group. Drawing upon life-history interviews with 18 former heroin drug dealers in an inland city in China, this article explores how the trajectories of the Chinese heroin generation were deeply embedded in and affected by macro social transitions, and how their collective trajectories configured the local illegal drug market. Influenced by the desire to break free from traditional constraints brought about by China's reform and opening-up, informants voluntarily left their legitimate work to pursue self-realization but eventually became lost in the rapid changes and turned to drug trafficking. During their involvement with heroin, they adhered to mainstream Chinese cultural norms: they entered the heroin market because of "face" and "fashion," and actively accumulated social, cultural, and criminal capital to operate their business. This article concludes that the heroin fad in China at the end of the last century is not so much a "moral crisis" as a reflection of the change in citizens' subjectivity and the unintended consequences of rapid economic development.

关键词： Latinas   Computer science   Undergraduate   Middle school   Counternarrative  

摘要： This study puts forth the counternarrative of the first author Gislaine, a first-generation undergraduate student, Latina, and computer science major. Gislaine participated in a research internship and STEM mentorship program led by the second author, Meredith. Through this program, Gislaine designed and taught CS lessons to predominantly low-income, African-American middle school students over an 8-day urban summer school program. By analyzing Gislaine's written account of her journey to CS, as well as planning documents and written reflections on her daily experiences mentoring youth, we explore how Gislaine's intersecting social identities shape her experiences both as a CS learner and as a mentor to youth. Gislaine's counternarrative illuminates the systems of oppression present in schools and the CS field, while also highlighting Gislaine's transformative approaches to mentoring students in CS. Gislaine's narratives emphasize the importance of humanizing CS and leveraging students' strengths to promote equitable access to CS education. Through her experiences, she underscores the significance of recognizing and addressing systemic barriers, while also advocating for inclusive and empowering educational practices.

关键词： Optimization   Approximation algorithms   Scheduling   Scheduling algorithms   Data communication   Computer science   Internet of Things   Age of information   data collection   directional charging   peak AoI   wireless-powered network edge  

摘要： Age of Information (AoI) has emerged as a new metric to measure data freshness from the destination's perspective. To optimize the system AoI, most existing works focused on the point of scheduling of update transmissions. While at wireless-powered network edge, the source nodes can only transmit their updates after being charged ready, which means the system AoI is not only determined by the update transmission strategies, but also the charging strategies. Thus, in this article, we investigate the first work to optimize the weighted peak AoI from the point of charging at wireless-powered network edge. First, the problem of optimizing the weighted sum of average peak AoI with a directional charger is formulated, and then transformed to a charging time optimization problem with respect to the charging orientations and peak AoI, and an approximate algorithm is proposed to obtain the required charging time for each source node. Second, an age-based scheduling algorithm is proposed to compute the charging decisions and transmission decisions simultaneously, which can not only optimize the weighted sum of average peak AoI, but also guarantee the maximum peak AoI of each source node is bounded. The proposed algorithm is proved to have an approximation ratio of up to (1+ phi ), where phi is a small value related to the weight of each source node. When there exist multiple chargers, an approximate algorithm is also proposed to minimize the weighted sum of average peak AoI by scheduling the orientations of these chargers cooperatively. Finally, the extensive simulations demonstrate the high performance of the proposed algorithms in terms of peak AoI.

关键词： Computer science   OLS regression   Community college transfer   PhD interest   Computer science majors  

摘要： PurposeSupporting community college transfer students represents a critical strategy for broadening participation in STEM. In addition to being a racially diverse group, students who pursue STEM degrees by way of community college report frequent interests in graduate study and academic careers. Thus, supporting and expanding transfer students' PhD interests can help to diversify the STEM professoriate. This study aims to identify the experiences that predict PhD interests among students who transferred into the computer science major from a community ***/methodology/approachRelying on longitudinal survey data from over 150 community college transfer students throughout their first year at their receiving four-year university, we used regression analysis to identify the post-transfer college experiences that predict early interest in *** found that receiving information about PhDs from a professor strongly predicted PhD interest among transfer students. Relationships with other variables indicate that the provision of information about graduate school was more likely to occur for students who participated in undergraduate research experiences than for those participating in internships. Descriptive data document inequities in who has access to these types of ***/valueThis paper provides new insight into how STEM departments can develop targeted efforts to ensure that information about PhD training is equitably available to all transfer students. Working to ensure that faculty equitably communicate with students about PhD opportunities may go a long way in countering potential deterrents among transfer students who may be interested in such pathways.

关键词： Computer science   Electrical and electronic engineering  

摘要： Reinforcement learning (RL) is an effective method in training dialogue policies to steer the conversation towards successful task completion. However, most RL-based methods only rely on semantic inputs that lack empathy as they ignore the user emotional information. Moreover, these methods suffer from delayed rewards caused by the user simulator returning valuable results only at dialogue end. Recently, some methods have been proposed to learn the reward function together with user emotions, but they omit considering user emotion in each dialogue turn. In this paper, we proposed an emotion-sensitive dialogue policy model (ESDP), it incorporates user emotions information into dialogue policy and selects the optimal action by the combination of top-k actions with the user emotions. The user emotion information in each turn is used as an immediate reward for the current dialogue state to solve sparse rewards and the dependency on termination. Extensive experiments validate that our method outperforms the baseline approaches when combined with different Q-Learning algorithms, and also surpasses other popular existing dialog policies' performance.

关键词： Medical imaging   Institutions   Computer science  

摘要： Radiological imaging is a globally prevalent diagnostic method, yet the free text contained in radiology reports is not frequently used for secondary purposes. Natural Language Processing can provide structured data retrieved from these reports. This paper provides a summary of the current state of research on Large Language Model (LLM) based approaches for information extraction (IE) from radiology reports. We conduct a scoping review that follows the PRISMA-ScR guideline. Queries of five databases were conducted on August 1st 2023. Among the 34 studies that met inclusion criteria, only pre-transformer and encoder-based models are described. External validation shows a general performance decrease, although LLMs might improve generalizability of IE approaches. Reports related to CT and MRI examinations, as well as thoracic reports, prevail. Most common challenges reported are missing validation on external data and augmentation of the described methods. Different reporting granularities affect the comparability and transparency of approaches.

关键词： Computer science   Scientific data  

摘要： Robust object detection is crucial for automating underwater marine debris collection. While supervised deep learning achieves state-of-the-art performance in discriminative tasks, replicating this success on underwater data is challenging. The generalization of these methods suffers due to a lack of available annotated data considering different sources of variation in the unstructured underwater environment and imaging conditions. In this paper, we present the Seaclear Marine Debris Dataset, the first publicly available shallow-water marine debris dataset annotated for instance segmentation/object detection. The dataset contains 8610 images collected using ROVs at multiple locations and with different cameras, annotated for 40 object categories, encompassing not only litter but also observed animals, plants, and robot parts. As part of the technical validation, we provide baseline results for object detection using Faster RCNN and YOLOv6 models. Furthermore, we demonstrate the non-triviality of generalizing the trained model performance to unseen sites and cameras due to domain shift. This underscores the value of the presented dataset in further developing robust models for underwater debris detection.

关键词： Climate change   Computer science   Energy   Machine learning   Sustainability  

摘要： Within our sample set, the most efficient question-answering model used approximately 0.1 watt-hours (roughly the energy needed to power a 25W incandescent light bulb for 5 minutes) to process 1,000 questions. The least efficient image-generation model, by contrast, required as much as 1,600 Wh to create 1,000 high-definition images - that's the power necessary to fully charge a smartphone approximately 70 times, amounting to a 16,000-fold difference. In general, supervised tasks such as question answering or text classification - in which models are provided with a set of options to choose from or a document that contains the answer - are much more energy efficient than are generative tasks that rely on the patterns learnt from the training data to produce a response from scratch3. [...]summarization and text-classification tasks use relatively little power, although it must be noted that nearly all use cases involving large language models are more energy intensive than a Google search (querying an AI chatbot once uses up about ten times the energy required to process a web search request). To make it standard practice to measure and publicly share the energy use of models (for example, in a 'model card' setting out information such as training data, evaluations of performance and metadata), it's essential to get developers on board.

关键词： Machine learning   Computer science  

摘要： When this process is repeated for several games, the network's performance on the first game might be reduced to the level of random button-pressing4. Because most Al systems are trained on fixed data sets, their inability to learn continually does not hamper the successful application of deep learning, a set of machine-learning algorithms that solve complex problems by processing data through 'deep' (multilayered) networks, in domains such as image recognition and language processing. Dohare etai, studied a pathology that is the complement of catastrophic forgetting: rather than losing capabilities as a result of training sequentially on multiple tasks, the neural network loses its ability to learn - in some cases becoming completely unable to improve its performance on new tasks. The test required the algorithm to continually update its learning - first training on images of cats and houses, for example, to recognize the difference between these objects, and then learning to distinguish stop signs from school buses, before moving on to the next task, and so on. Because the difficulty of tasks did not increase, a loss of learning ability was implicit in the reduced accuracy with which a standard network classified the images as new tasks were added (Fig. 1).