We propose a novel framework that leverages Visual Question Answering (VQA) models to automate the evaluation of LLM-generated data visualizations. Traditional evaluation methods often rely on human judgment, which is costly and unscalable, or focus solely on data accuracy, neglecting the effectiveness of visual communication. By employing VQA models, we assess data representation quality and the general communicative clarity of charts. Experiments were conducted using two leading VQA benchmark datasets, ChartQA and PlotQA, with visualizations generated by OpenAI’s GPT-3.5 Turbo and Meta’s Llama 3.1 70B-Instruct models. Our results indicate that LLM-generated charts do not match the accuracy of the original non-LLM-generated charts based on VQA performance measures. Moreover, while our results demonstrate that few-shot prompting significantly boosts the accuracy of chart generation, considerable progress remains to be made before LLMs can fully match the precision of human-generated graphs. This underscores the importance of our work, which expedites the research process by enabling rapid iteration without the need for human annotation, thus accelerating advancements in this field.

Sensor networks may be used in many monitoring applications where the locations of the monitored objects are quite sensitive and need to be protected. Previous research mainly focuses on protecting source location against mote-class attackers who only have a local view of the network traffic. In this paper, we focus on how to protect the source location against laptop-class attackers who have a global view of the network traffic. This paper proposes four schemes---naive, global, greedy, and probabilistic---to deal with laptop-class attacks. The naive solution uses maintenance messages sent periodically to hide real event reports. The global and greedy solutions improve the naive solution by reducing the latency of event delivery without increasing communication overhead. The probabilistic solution further improves the performance by reducing communication overhead without sacrificing location privacy. Experiments show that the probabilistic solution is practical for providing source location privacy against a laptop-class attacker.

Localization is a fundamental problem in wireless sensor networks. Currently, most localization protocols assume a benign environment. This paper studies how a localization protocol can be affected by attacks, especially those like wormhole attacks that don't need to capture the keys used in the network, and how such an attack can be detected and defended in distributed scheme. By applying several localization algorithms, this paper first shows that wormhole attacks are a serious threat to all these algorithms, based on simulation in NS-2. The paper first evaluates the effects of wormhole attacks on several Range-Free Anchor-Free (RFAF) localization algorithms based on hop-counting and connectivity techniques. Then, the paper presents an RFAF localization algorithm called Wormhole-resilient Geographic Distributed Localization (WGDL), which embeds a wormhole detecting/recovering mechanism: if this mechanism detects wormhole attacks in location computation, then it will restore the localization by freezing the wormhole affected area. Simulations show that the proposed detection method is effective on different network placements, and that the mechanism has both a low False Toleration Rate (FTR) and a low False Detection Rate (FDR) in detecting wormhole attacks.

Sensor networks are used in a variety of applications such as battlefield reconnaissance, environmental monitoring, and traffic monitoring. Security and privacy become important concerns when people are participants in sensor network applications [1]. In this research, we focus on providing a framework in which the privacy of people participating in sensor networks applications is protected.

Localization is a fundamental problem in sensor networks. This paper proposes a new distributed localization algorithm called GDL ("Geographic Distributed Localization") that is based on hop-counting. Although the idea of predicting node position from connectivity information in WSNs is not new, our algorithm improves on other localization algorithms of this type in several key ways. Results from extensive simulation in NS-2 show a significant improvement in the accuracy with which locations can be determined in WSNs with varying density and different number of nodes on irregular placements. In particular, when compared with the MDS-MAP series of algorithms, our algorithm achieves about a 30-40% improvement in accuracy. It also has a lower communication cost (0(n) versus O(nlogn)) than MDS-MAP(P), and has a low constant memory cost per node.

Recent research has identified significant vulnerabilities in recommender systems. Shilling attacks, in which attackers introduce biased ratings in order to influence future recommendations, have been shown to be effective against collaborative filtering algorithms. We postulate that the distribution of item ratings in time can reveal the presence of a wide range of shilling attacks given reasonable assumptions about their duration. To construct a time series of ratings for an item, we use a window size of k to group consecutive ratings for the item into disjoint windows and compute the sample average and sample entropy in each window. We derive a theoretically optimal window size to best detect an attack event if the number of attack profiles is known. For practical applications where this number is unknown, we propose a heuristic algorithm that adaptively changes the window size. Our experimental results demonstrate that monitoring rating distributions in time series is an effective approach for detecting shilling attacks.

Collaborative filtering techniques have become popular in the past several years as an effective way to help people deal with information overload. An important security concern in traditional recommendation systems is that users disclose information that may compromise their individual privacy when providing ratings. Randomized perturbation schemes have been proposed to disguise user ratings while still producing accurate recommendations. However, recent research has suggested that perturbation schemes might not be able to preserve privacy as much as has been believed. We propose two data reconstruction methods that derive original private information from disguised data in existing perturbation collaborative filtering schemes. One method is based on k-means clustering and the other uses singular value decomposition (SVD). We have conducted theoretical and experimental analysis on the difference between original data and reconstructed data. Our experiments show that both methods can derive a considerable amount of original information. This study helps to determine an empirical trade-off between recommendation accuracy and user privacy in perturbation schemes.

We use a low-dimensional linear model to describe the user rating matrix in a recommendation system. A non-negativity constraint is enforced in the linear model to ensure that each user's rating profile can be represented as an additive linear combination of canonical coordinates. In order to learn such a constrained linear model from an incomplete rating matrix, we introduce two variations on Non-negative Matrix Factorization (NMF): one based on the Expectation-Maximization (EM) procedure and the other a Weighted Nonnegative Matrix Factorization (WNMF). Based on our experiments, the EM procedure converges well empirically and is less susceptible to the initial starting conditions than WNMF, but the latter is much more computationally efficient. Taking into account the advantages of both algorithms, a hybrid approach is presented and shown to be effective in real data sets. Overall, the NMF-based algorithms obtain the best prediction performance compared with other popular collaborative filtering algorithms in our experiments; the resulting linear models also contain useful patterns and features corresponding to user communities.

An important security concern with traditional recommendation systems is that users disclose information that may compromise their individual privacy when providing ratings. A randomization approach has been proposed to disguise user ratings while still producing accurate recommendations. However, recent research has suggested that a significant amount of original private information can be derived from perturbed data in a randomization scheme. We suggest that a main limitation of the existing randomization approach is that perturbation is item-invariant--each item has a same perturbation variance. Based on this observation, we introduce a two-way communication privacypreserving scheme in which users perturb their ratings for each item based on the server's guidance instead of using an item-invariant perturbation. Compared to the existing randomization approach, our new scheme can help users disclose much less private information at the same recommendation accuracy level.

Collaborative filtering techniques have become popular in the past decade as an effective way to help people deal with information overload. Recent research has identified significant vulnerabilities in collaborative filtering techniques. Shilling attacks, in which attackers introduce biased ratings to influence recommendation systems, have been shown to be effective against memory-based collaborative filtering algorithms. We examine the effectiveness of two popular shilling attacks (the random attack and the average attack) on a model-based algorithm that uses Singular Value Decomposition (SVD) to learn a low-dimensional linear model. Our results show that the SVD-based algorithm is much more resistant to shilling attacks than memory-based algorithms. Furthermore, we develop an attack detection method directly built on the SVD-based algorithm and show that this method detects random shilling attacks with high detection rates and very low false alarm rates.

Sensor networks are used in a variety of application areas for diverse problems from habitat monitoring to military tracking. Whenever they are used to monitor sensitive objects, the privacy of monitored objects' locations becomes an important concern. When a sensor reports a monitored object by sending a series of messages through the sensor network, the route these messages take in theory creates a trail leading back to their source. By eavesdropping on communications, an attacker may be able to move from node to node to follow this trail. Several approaches aimed at discouraging this kind of eavesdropping have been proposed, including mechanisms for constructing "phantom" routes and approaches that insert fake sources as background noise. A problem with existing approaches is that message latencies become larger and energy costs become higher as a result of introducing protections for the privacy of a source location. This paper proposes a new cyclic entrapment method (CEM) that protects source locations in sensor networks while adding a comparatively low cost in terms of additional message latency and energy.

The increasing popularity of peer-to-peer systems has promoted the development of new techniques to support various kinds of business transactions. However, users are reluctant to conduct high value transactions over P2P systems due to the inherent untrustworthiness of peers. In this paper, we investigate building trust through automated trust negotiation, which is orthogonal to existing reputation-based approaches. Trust negotiation makes it possible to prove a peer satisfies certain trust requirements imposed by the ongoing business. We also introduce locally trusted third parties, which we demonstrate can effectively promote successful trust negotiations in peer-to-peer systems.

This paper describes a hybrid negotiation strategy mechanism using a strategy pool framework that allows negotiation agents more flexibility and robustness in an automated negotiation system. The strategy pool framework supports: a) dynamically assigning an appropriate negotiation strategy to a negotiation agent according to the current negotiation environment and b) creating new negotiation rules by learning from past negotiations. Learning forms we use here for the framework are Feed Forward Back Propagation (FFBP) neural networks and multidimensional inter-transaction association rules mining.

msmcT Functional hia~etic Resonance tiagery (~~) is a new' medical ima=tig technology pro~ti~mg tictional, as opposed to natomicd, mapptig of the human brain.The promise of this non-invasive technique has opened new' ch~enges for mdtimedia researchers.A the fieId of human functional neurotia=tig e~lodes and the technology of ~W advmces, computational techniques for expertient control, subject stimu~ generation, and joint stimti-activation tiaetig are Iachg.host computational work h= focused on impro~tig the analysis of the brab S=S.Tti paper describes eomputiationa~mechanisms for the dehery md tiactig of multimedia stimti and a correlation framework for stimti and brain responses.NIediaStim, a new integrated data co~ection fraework for stimuIus tracking $hat k currently under development. NlediaSfi etiends the stimti paradi=m to composite multimedia presentations. To stiulate a real-tie sitiation as reafitica~y as possible, it is tiportant to stidy the brainPerrnksion lo ma}:edigi~alor hard copies of fl or part of thk ~!;ork for personal or classroom use k granted ;I:ithout fee pro~,ided ~ha~copies are not made or distributed for profit or commertizl ada~antage, and fiat copies bear this no:i~and ~heftil titation on the firs!page.To copy oihen!.k~lo repub~i~to post on senrers or 10 redistribute 10 lis~, requires prior specific perrnksion anifor a fee.