Entire Dataset Research Articles

Once and for All: Universal Transferable Adversarial Perturbation against Deep Hashing-Based Facial Image Retrieval

Deep Hashing (DH)-based image retrieval has been widely applied to face-matching systems due to its accuracy and efficiency. However, this convenience comes with an increased risk of privacy leakage. DH models inherit the vulnerability to adversarial attacks, which can be used to prevent the retrieval of private images. Existing adversarial attacks against DH typically target a single image or a specific class of images, lacking universal adversarial perturbation for the entire hash dataset. In this paper, we propose the first universal transferable adversarial perturbation against DH-based facial image retrieval, a single perturbation can protect all images. Specifically, we explore the relationship between clusters learned by different DH models and define the optimization objective of universal perturbation as leaving from the overall hash center. To mitigate the challenge of single-objective optimization, we randomly obtain sub-cluster centers and further propose sub-task-based meta-learning to aid in overall optimization. We test our method with popular facial datasets and DH models, indicating impressive cross-image, -identity, -model, and -scheme universal anti-retrieval performance. Compared to state-of-the-art methods, our performance is competitive in white-box settings and exhibits significant improvements of 10%-70% in transferability in all black-box settings.

Divide-and-Aggregate Learning for Evaluating Performance on Unlabeled Data

Artificial Intelligence (AI) models have become an integral part of modern society, significantly improving human lives. However, ensuring the reliability and safety of these models is of paramount importance. One critical aspect is the continuous monitoring and verification of model performance to prevent any potential risks. Real-time online evaluation of AI models is necessary to maintain their effectiveness and mitigate any harm caused by performance degradation. The traditional approach to model evaluation involves supervised methods that rely on manual labeling to compare results with model predictions. Unfortunately, this method is not suitable for online model monitoring due to its inherent lag and high cost. While there have been attempts to explore free-label model evaluation, these approaches often consider only the global features of the entire dataset. Additionally, they can only perform model evaluation based on a single dimension of model confidence or features. In this paper, we propose a novel approach called Divide-and-Aggregate Learning (DAL) for unsupervised model evaluation. Our method addresses the limitations of previous approaches by dividing the output of the model into buckets, capturing local information of the distribution. We then aggregate this local information to obtain global information and further represent the relationship between the distribution and model performance. Importantly, our method can simultaneously handle the confidence distribution and feature distribution of the model output. Extensive experiments have been conducted to demonstrate the effectiveness of our DAL model. The results show that our approach outperforms previous methods on four widely used datasets. We will make our source code publicly available.

Parameterized Approximation Algorithms for Sum of Radii Clustering and Variants

Clustering is one of the most fundamental tools in artificial intelligence, machine learning, and data mining. In this paper, we follow one of the recent mainstream topics of clustering, Sum of Radii (SoR), which naturally arises as a balance between the folklore k-center and k-median. SoR aims to determine a set of k balls, each centered at a point in a given dataset, such that their union covers the entire dataset while minimizing the sum of radii of the k balls. We propose a general technical framework to overcome the challenge posed by varying radii in SoR, which yields fixed-parameter tractable (fpt) algorithms with respect to k (i.e., whose running time is f(k) ploy(n) for some f). Our framework is versatile and obtains fpt approximation algorithms with constant approximation ratios for SoR as well as its variants in general metrics, such as Fair SoR and Matroid SoR, which significantly improve the previous results.

Task offloading in mobile edge computing using cost-based discounted optimal stopping

Abstract Mobile edge computing (MEC) paradigm has emerged to improve the quality of service & experience of applications deployed in close proximity to end-users. Due to their restricted computational and communication resources, MEC nodes can provide access to a portion of the entire set of services and data gathered. Therefore, there are several obstacles to their management. Keeping track of all the services offered by the MEC nodes is challenging, particularly if their demand rates change over time. Received tasks (such as, analytics queries, classification tasks, and model learning) require services to be invoked in real MEC use-case scenarios, e.g., smart cities. It is not unusual for a node to lack the necessary services or part of them. Undeniably, not all the requested services may be locally available; thus, MEC nodes must deal with the timely and appropriate choice of whether to carry out a service replication (pull action) or tasks offloading (push action) to peer nodes in a MEC environment. In this study, we contribute with a novel time-optimized mechanism based on the optimal stopping theory, which is built on the cost-based decreasing service demand rates evidenced in various service management situations. Our mechanism tries to optimally solve the decision-making dilemma between pull and push action. The experimental findings of our mechanism and its comparative assessment with other methods found in the literature showcase the achieved optimal decisions with respect to certain cost-based objective functions over dynamic service demand rates.

Abstract 2959: Signatures of therapy-induced senescence in glioblastoma-derived extracellular vesicles

Abstract The recalcitrant nature of glioblastoma (GBM) may be due, in part, to the promotion of therapy-induced senescence (TIS) by standard of care treatments. TIS is a prolonged growth arrest associated with epigenetic, metabolic, and biochemical alterations, as well as a robust secretory phenotype. Despite an initial period of tumor stasis due to the loss of cell proliferation, TIS is associated with the development of stemness, increased drug resistance, and immune suppression. Our preliminary data demonstrate the prominent induction of TIS in GBM tumor cells and astrocytes following standard of care therapies for GBM in vitro. However, studies of TIS in GBM patients have been limited due to i) the impracticality of receiving multiple GBM biopsies post-treatment but prior to recurrence and ii) the lack of robust detection methods for TIS even with appropriate biopsy. These challenges argue the need for novel methods to survey TIS in patients. Extracellular vesicles (EVs) are small membrane-bound particles that readily cross the blood brain barrier, and convey information about originating cell identity and cell state through their cargo content. We hypothesize that TIS in host and tumor cells drives the elaboration of senescence-associated EVs (senEVs) with quantifiable senescence protein cargo signatures, and consequently, that EVs can serve as a liquid biopsy for senescence in patients. In a preliminary experiment, two patient derived xenografts grown as short-term in vitro cultures were exposed to 6GY irradiation, with supernatant collected prior to radiation (D0), at peak senescence (D7), and at late senescence (D10). EVs were isolated and subjected to unbiased mass spectrometry. Following unsupervised clustering, EVs were grouped by treatment rather than by cell line. Approximately 6000 proteins were identified across the entire data set, with 105 proteins unique to D7 samples, 162 proteins unique to D10 samples, and 688 proteins exclusive to D7 and D10 samples. Further, proteins from two established senescence gene sets were identified as unique to the D7 and/or D10 conditions. Current work to more robustly identify enrichment of senescence gene sets in EV cargo post-irradiation is underway. We have also established shRNA knockdown cell lines with abrogated TIS to equip validation studies of the TIS-EV cargo relationship. We anticipate that these studies will aid in the development of EVs as a novel reporter of senescence in the clinic. Citation Format: Valerie DeLuca, Priya Digumarti, Krystine Garcia-Mansfield, Patrick Pirrotte, Michael E. Berens. Signatures of therapy-induced senescence in glioblastoma-derived extracellular vesicles [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 2959.

Abstract 2584: Decoding colon cancer recurrence: Unveiling accurate predictions with attention-guided deep neural networks on histopathological whole slide images

Abstract Background: Up to 40% of colon cancer patients are at high risk of cancer recurrence, yet accurate and timely prediction tools are lacking. Leveraging whole slide images (WSIs) and deep learning models, we aimed to develop precise algorithm for prediction of colon cancer recurrence. Thus, enables risk stratification for optimized therapeutic interventions and improved health outcomes. Design: We designed an attention-based deep learning model for predicting colon cancer recurrence on paraffin-embedded, hematoxylin and eosin-stained colon tissue biopsies of digital slides. The WSIs were downloaded from the TCGA dataset, and preprocessed for color normalization, tissue segmentation, tilling, and histopathological features extraction. The entire dataset was then labeled based on cancer recurrence status, and divided into training (70%), validation (15%), and testing sets (15%). The model's performance was then evaluated by standard evaluation metrics, including receiver operating characteristic Area Under the Curve (AUC) and accuracy, on the training, validation, and testing datasets, where interpretability attention heatmaps were applied to gain insights into specific histological features and patterns involved in the model's decision-making process. The study is supported by the NIH-NCI-T32 for Next Generation Pathologists Program at our institution. Results: A total of 350 WSIs were included and labeled into two classes: post-therapeutic colon cancer recurrence and no recurrence. The tissue segmentation process involved converting RGB images to HSV color space, applying a median blur filter (kernel size: 7), and performing thresholding using Otsu's method. The extracted features were utilized to train and construct a Clustering-constrained Attention Multiple Instance Learning (CLAM) model. The model demonstrated consistent performance across the validation and testing datasets, achieving an AUC of 0.85 with an accuracy of 0.83 on the validation set, and an identical AUC of 0.85 with an accuracy of 0.83 on the testing set, indicating the model's robust ability to identify patients at risk of cancer recurrence. Conclusion: The study demonstrates the strong performance of our deep learning model in accurately identifying patients at risk for colon cancer recurrence based on H&E WSIs. This capability paves the way for optimizing therapeutic interventions and implementing effective surveillance strategies. Consequently, highlighting the crucial role of pathologists in collaborating with the oncologist for optimizing the management of colon cancer care in the era of personalized medicine. Citation Format: Mohammad K. Alexanderani, Mohamed Omar, Matthew Greenblatt, Ethel Cesarman, Maria T. Diaz-Meco, Jorge Moscat, Luigi Marchionni. Decoding colon cancer recurrence: Unveiling accurate predictions with attention-guided deep neural networks on histopathological whole slide images [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 2584.

Abstract 5479: Spatially resolved molecular features in the tumor-immune microenvironment of germline BRCA1/2 mutated versus BRCA wild type high grade serous ovarian cancers

Abstract High-grade serous ovarian cancer (HGSC) is the most common and aggressive subtype of epithelial ovarian cancer. Importantly, germline mutations in BRCA1/2 give rise to HGSC. Because most HGSC are diagnosed at stage III/IV, outcomes are poor and more effective therapies are needed. Although immune checkpoint inhibitors show promise for a number of cancer types, efficacy has been variable in HGSC. To help provide new clues to the molecular features of HGSC, we set out to elucidate the tumor microenvironment of HGSC using spatial transcriptomics approaches. Specifically, we applied 10x Genomics Visium Spatial Transcriptomics (ST) technology to 48 samples representing primary HGSC tumors from 24 patients, including 15 BRCA1/2 mutated cases and 9 BRCA non-mutated cases. In addition to individual sample analysis, we normalized data from 64,784 total features across all sections and performed an integrated cluster analysis to get an overall picture of the cellular architecture of HGSC, but to also directly investigate differences in the tumor microenvironment based on BRCA mutation status. The resulting 17 integrated clusters were subjected to molecular annotation using GSEA, ESTIMATE, and CIBERSORTx, which revealed feature clusters representing important cell types defined by marker genes. Analysis of the entire dataset identified a primary tumor cluster defined by the expression of known ovarian cancer genes including WFDC2 and “don’t eat me” marker genes CD24 and CD47. However, the predominant tumor cell population in BRCA mutated cases was marked by the expression of MUC1, EPCAM, CLDN3, and CD24, while BRCA non-mutated tumor cell cluster was marked by EFNA1, ID1, LSR, and CD9. Importantly, both groups revealed the presence of a specialized monocyte derived macrophage cluster defined by the expression of LYZ and genes involved in cholesterol metabolism including APOE and APOC1. These lipid-associated macrophages are M2 polarized and shown to be associated with poor outcome. Additional analyses are underway to determine spatial dependencies of various tumor and immunological cell compartments. These data provide novel insights into the HGSC tumor microenvironment that may influence HGSC’s vulnerability to immunological therapies. Citation Format: Jing Qian, Lynda D. Roman, Rania Bassiouni, Seeta Rajpara, Monica Neuman, Varun Khetan, David W. Craig, Joseph W. Carlson, John D. Carpten. Spatially resolved molecular features in the tumor-immune microenvironment of germline BRCA1/2 mutated versus BRCA wild type high grade serous ovarian cancers [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 5479.

The Research on the Improvement of FP-growth Algorithm

This paper focuses on the issue of low efficiency in the FP-growth algorithm for frequent pattern mining and proposes an improved algorithm, ICFM-growth. Experimental results demonstrate that the improved algorithm outperforms the FP-growth algorithm in terms of both runtime and space utilization. By studying the classical frequent pattern algorithms Apriori and FP-growth, the latter is an improved algo- rithm based on Apriori to address the problems of generating a large number of candidate itemsets and consuming significant memory space. While FP-growth exhibits superior mining efficiency compared to Apriori, it faces challenges when dealing with large and long transactional databases due to the construction of numerous FP-trees, which increases computational tasks and prolongs runtime, leading to lagging mining efficiency. To address this issue, the improved algorithm ICFM-growth is proposed. It constructs a co-occurrence frequency matrix to perform preliminary screening on the transaction set, focusing on high-frequency and co-occurring item pairs, thereby reducing unnecessary computa- tions. In the initial stage, as important item pairs have already been filtered, the algorithm can directly operate on these item pairs and items, rather than the entire dataset, thereby reducing the search space and computational complexity. Additionally, the structure of the FP-tree ena- blesthe algorithm to store and process data more efficiently, avoiding frequent scanning of the entire database, as seen in traditional Apriori algorithms. Finally, through simulation experiments on publicly available datasets such as Movie Data and House Data, validated using cross- validation, the ICFM-growth algorithm proves to be significantly superior to FP-growth in terms of time and space efficiency. It demonstrates faster runtime, lower memory consumption, and superior mining efficiency compared to FP-growth.

Introducing diminutive causal structure into graph representation learning

When engaging in end-to-end graph representation learning with Graph Neural Networks (GNNs), the intricate causal relationships and rules inherent in graph data pose a formidable challenge for the model in accurately capturing authentic data relationships. A proposed mitigating strategy involves the direct integration of rules or relationships corresponding to the graph data into the model. However, within the domain of graph representation learning, the inherent complexity of graph data obstructs the derivation of a comprehensive causal structure that encapsulates universal rules or relationships governing the entire dataset. Instead, only specialized diminutive causal structures, delineating specific causal relationships within constrained subsets of graph data, emerge as discernible. Motivated by empirical insights, it is observed that GNN models exhibit a tendency to converge towards such specialized causal structures during the training process. Consequently, we posit that the introduction of these specific causal structures is advantageous for the training of GNN models. Building upon this proposition, we introduce a novel method that enables GNN models to glean insights from these specialized diminutive causal structures, thereby enhancing overall performance. Our method specifically extracts causal knowledge from the model representation of these diminutive causal structures and incorporates interchange intervention to optimize the learning process. Theoretical analysis serves to corroborate the efficacy of our proposed method. Furthermore, empirical experiments consistently demonstrate significant performance improvements across diverse datasets.

CYTOKINE PROFILE RESPONSE IN PATIENTS WITH FRACTURES OF LONG BONES DURING COVID-19

Introduction: The immune system swiftly responds to the onslaught of trauma and the infectious agents it triggers. However, the mechanisms of action of the COVID-19 virus remain unidentified to date. Distinctive alterations that occur in the immune system of trauma victims with COVID-19, setting them apart from changes observed in other diseases. This underscores the specificity of traumatic disease within this patient with COVID-19.
 The study aims to assess the response of humoral immunity markers in patients with fractures of long bones and COVID-19 and to explore their role in early diagnosing infectious complications in this particular group of patients.
 Materials and methods: To accomplish the study objectives, we assembled a retrospective dataset comprising 289 cases of skeletal bone fractures treated at the Kyiv City Clinical Hospital of Emergency Medical Care from March, 2020 to February, 2021, meeting the inclusion criteria for the study. The entire dataset consisted of 289 cases of skeletal bone fractures, categorized into two groups: the main group and the control group.
 Results: On the first day of treatment, an increase in some pro-inflammatory cytokines (IL-6, IL-1β and IL-4) with a normal level of TNF-α, normal values of IFN-γ, IFN-α and a sharp decrease in the anti-inflammatory cytokine IL was observed among the patients of the main group -10. On the third day of the treatment, there was a significant increase in all pro-inflammatory cytokines (IL-6, IL-1β, IL-4 and TNF-α), both types of interferons, and an even sharper decrease in the anti-inflammatory cytokine IL-10. On the tenth day, the growth in the level of pro-inflammatory cytokines was noted, among which the level of TNF-α was the leader, a subnormal level of interferons and anti-inflammatory interleukin 10 was observed. Conclusions: The dynamics of pro-inflammatory cytokines in patients with long bone fractures against the background of COVID-19 tended to increase and had a maximum level on the third day of treatment. Anti-inflammatory cytokines demonstrate a tendency to depression and reache the reference value only on the 10th day of the study.

Cost-aware retraining for machine learning

Retraining a machine learning (ML) model is essential for maintaining its performance as the data change over time. However, retraining is also costly, as it typically requires re-processing the entire dataset. As a result, a trade-off arises: on the one hand, retraining an ML model too frequently incurs unnecessary computing costs; on the other hand, not retraining frequently enough leads to stale ML models and incurs a cost in loss of accuracy. To resolve this trade-off, we envision ML systems that make automated and cost-optimal decisions about when to retrain an ML model.In this work, we study the decision problem of whether to retrain or keep an existing ML model based on the data, the model, and the predictive queries answered by the model. Crucially, we consider the costs associated with each decision and aim to optimize the trade-off. Our main contribution is a Cost-Aware Retraining Algorithm, CARA, which optimizes the trade-off over streams of data and queries. To explore the performance of CARA, we first analyze synthetic datasets and demonstrate that CARA can adapt to different data drifts and retraining costs while performing similarly to an optimal retrospective algorithm. Subsequently, we experiment with real-world datasets and demonstrate that CARA has better accuracy than drift detection baselines while making fewer retraining decisions, thus incurring lower total costs.

Developing a New ANN Model to Estimate Daily Actual Evapotranspiration Using Limited Climatic Data and Remote Sensing Techniques for Sustainable Water Management

Accurate estimations of actual evapotranspiration (ETa) are essential to various environmental issues. Artificial intelligence-based models are a promising alternative to the most common direct ETa estimation techniques and indirect methods by remote sensing (RS)-based surface energy balance models. Artificial Neural Networks (ANNs) are proven to be suitable for predicting reference evapotranspiration (ETo) and ETa based on RS data. This study aims to develop a methodology based on ANNs for estimating daily ETa values using NDVI and land surface temperature, coupled with limited site-specific climatic variables in a large irrigation catchment. The ANN model has been applied to the two different scenarios. Data from only the 38 days of satellite overpass dates was selected in Scenario I, while in Scenario II all datasets, i.e., the 769-day data were used. An irrigation scheme, located in the Mediterranean region of Turkiye, was selected, and a total of 38 Landsat images and local climatic data collected in 2021 and 2022 were used in the ANN model. The ETa results by the ANN model for Scenarios I and II showed that the R2 values for training (0.79 and 0.86), testing (0.75 and 0.81), and the entire dataset (0.76 and 0.84) were all remarkably high. Moreover, the results of the new ANN model in two scenarios showed an acceptable agreement with ETa-METRIC values. The proposed ANN model demonstrated the potential for obtaining daily ETa using limited climatic data and RS imagery. As a result, the suggested ANN model for daily ETa computation offers a trustworthy way to determine crop water usage in real time for sustainable water management in agriculture. It may also be used to assess how crop evapotranspiration in drought-prone areas will be affected by climate change in the 21st century.

NuInsSeg: A fully annotated dataset for nuclei instance segmentation in H&E-stained histological images

In computational pathology, automatic nuclei instance segmentation plays an essential role in whole slide image analysis. While many computerized approaches have been proposed for this task, supervised deep learning (DL) methods have shown superior segmentation performances compared to classical machine learning and image processing techniques. However, these models need fully annotated datasets for training which is challenging to acquire, especially in the medical domain. In this work, we release one of the biggest fully manually annotated datasets of nuclei in Hematoxylin and Eosin (H&E)-stained histological images, called NuInsSeg. This dataset contains 665 image patches with more than 30,000 manually segmented nuclei from 31 human and mouse organs. Moreover, for the first time, we provide additional ambiguous area masks for the entire dataset. These vague areas represent the parts of the images where precise and deterministic manual annotations are impossible, even for human experts. The dataset and detailed step-by-step instructions to generate related segmentation masks are publicly available on the respective repositories.

Precise atom-to-atom mapping for organic reactions via human-in-the-loop machine learning

Atom-to-atom mapping (AAM) is a task of identifying the position of each atom in the molecules before and after a chemical reaction, which is important for understanding the reaction mechanism. As more machine learning (ML) models were developed for retrosynthesis and reaction outcome prediction recently, the quality of these models is highly dependent on the quality of the AAM in reaction datasets. Although there are algorithms using graph theory or unsupervised learning to label the AAM for reaction datasets, existing methods map the atoms based on substructure alignments instead of chemistry knowledge. Here, we present LocalMapper, an ML model that learns correct AAM from chemist-labeled reactions via human-in-the-loop machine learning. We show that LocalMapper can predict the AAM for 50 K reactions with 98.5% calibrated accuracy by learning from only 2% of the human-labeled reactions from the entire dataset. More importantly, the confident predictions given by LocalMapper, which cover 97% of 50 K reactions, show 100% accuracy for 3,000 randomly sampled reactions. In an out-of-distribution experiment, LocalMapper shows favorable performance over other existing methods. We expect LocalMapper can be used to generate more precise reaction AAM and improve the quality of future ML-based reaction prediction models.

Mapping of sea ice concentration using the NASA NIMBUS 5 Electrically Scanning Microwave Radiometer data from 1972–1977

Abstract. The Electrically Scanning Microwave Radiometer (ESMR) instrument onboard the NIMBUS 5 satellite was a one-channel microwave radiometer that measured the 19.35 GHz horizontally polarized brightness temperature (TB) from 11 December 1972 to 16 May 1977. The original tape archive data in swath projection have recently been made available online by the NASA Goddard Earth Sciences Data and Information Services Center (GES DISC). Even though the ESMR was a predecessor of modern multi-frequency radiometers, there are still parts of modern processing methodologies which can be applied to the data to derive the sea ice extent globally. Here, we have reprocessed the entire dataset using a modern processing methodology that includes the implementation of pre-processing filtering, dynamical tie points, and a radiative transfer model (RTM) together with numerical weather prediction (NWP) for atmospheric correction. We present the one-channel sea ice concentration (SIC) algorithm and the model for computing temporally and spatially varying SIC uncertainty estimates. Post-processing steps include resampling to daily grids, land-spillover correction, the application of climatological masks, the setting of processing flags, and the estimation of sea ice extent, monthly means, and trends. This sea ice dataset derived from the NIMBUS 5 ESMR extends the sea ice record with an important reference from the mid-1970s. To make it easier to perform a consistent analysis of sea ice development over time, the same grid and land mask as used for EUMETSAT's OSI-SAF SMMR-based sea-ice climate data record (CDR) were used for our ESMR dataset. SIC uncertainties were included to further ease comparison to other datasets and time periods. We find that our sea ice extent in the Arctic and Antarctic in the 1970s is generally higher than those available from the National Snow and Ice Data Center (NSIDC) Distributed Active Archive Center (DAAC), which were derived from the same ESMR dataset, with mean differences of 240 000 and 590 000 km2, respectively. When comparing monthly sea ice extents, the largest differences reach up to 2 million km2. Such large differences cannot be explained by the different grids and land masks of the datasets alone and must therefore also result from the differences in data filtering and algorithms, such as the dynamical tie points and atmospheric correction. The new ESMR SIC dataset has been released as part of the ESA Climate Change Initiative (ESA CCI) program and is publicly available at https://doi.org/10.5285/34a15b96f1134d9e95b9e486d74e49cf (Tonboe et al., 2023).

SafeCheck: Detecting smart contract vulnerabilities based on static program analysis methods

AbstractEthereum smart contracts are a special type of computer programs. Once deployed on the blockchain, they cannot be modified. This presents a significant challenge to the security of smart contracts. Previous research has proposed static and dynamic detection tools to identify vulnerabilities in smart contracts. These tools check contract vulnerabilities based on predefined rules, and the accuracy of detection strongly depends on the design of the rules. However, the constant emergence of new vulnerability types and strategies for vulnerability protection leads to numerous false positives and false negatives by tools. To address this problem, we analyze the characteristics of vulnerabilities in smart contracts and the corresponding protection strategies. We convert the contracts' bytecode into an intermediate representation to extract semantic information of the contracts. Based on this semantic information, we establish a set of detection rules based on semantic facts and implement a vulnerability detection tool SafeCheck using static program analysis methods. The tool is used to detect six common types of vulnerabilities in smart contracts. We have extensively evaluated SafeCheck on real Ethereum smart contracts and compared it to other tools. The experimental results show that SafeCheck performs better in smart contract vulnerability detection compared to other typical tools, with a high F‐measure (up to 83.1%) for its entire dataset.

Anomaly Detection Algorithms for Low-Dimensional and High-Dimensional Data: A Critical Study

Suspicious events or objects that differ from the norm in data can be discovered using anomaly identification. Identifying anomalies is critical for many applicable domains of life, e.g., preventing credit card theft and spotting intrusions into networks. It is possible to detect anomalies on a global scale as well as at the local level. A global outlier is a data point beyond the norm for the entire dataset, while a local outlier may be inside the norm for the entire dataset but outside the surrounding data points. Data outlier identification methods that are performed locally are inadequate. Therefore, better algorithms are required to investigate the high velocity of data and identify local outliers. Machine learning and data mining techniques need to be investigated to determine the pros and cons of anomaly identification residing inside data. The density based LOF method can be applied as the best choice to identify local outliers. While many variants of LOF exist for low-dimensional data, none are suitable for high-dimensional data. This research study discusses LOF, COF, and CBLOF methods for spotting local outliers in low and high-dimensional data. Regarding the size of the dimension, the performance of density-based algorithms is examined based on accuracy and time complexity. In this scenario, CBLOF achieves outstanding results due to its distinctive method of employing cluster-based local outlier detection.

QSPR for the prediction of critical micelle concentration of different classes of surfactants using machine learning algorithms

The determination of the critical micelle concentration (CMC) is a crucial factor when evaluating surfactants, making it an essential tool in studying the properties of surfactants in various industrial fields. In this present research, we assembled a comprehensive set of 593 different classes of surfactants including, anionic, cationic, nonionic, zwitterionic, and Gemini surfactants to establish a link between their molecular structure and the negative logarithmic value of critical micelle concentration (pCMC) utilizing quantitative structure-property relationship (QSPR) methodologies. Statistical analysis revealed that a set of 14 significant Mordred descriptors (SlogP, GATS6d, nAcid, GATS8dv, GATS4dv, PEOE_VSA11, GATS8d, ATS0p, GATS1d, MATS5p, GATS3d, NdssC, GATS6dv and EState_VSA4), along with temperature, served as appropriate inputs. Different machine learning methods, such as multiple linear regression (MLR), random forest regression (RFR), artificial neural network (ANN), and support vector regression (SVM), were employed in this study to build QSPR models. According to the statistical coefficients of QSPR models, SVR with Dragonfly hyperparameter optimization (SVR-DA) was the most accurate in predicting pCMC values, achieving (R2 = 0.9740, Q2 = 0.9739, r‾m2 = 0.9627, and Δrm2 = 0.0244) for the entire dataset.

Density peak clustering by local centers and improved connectivity kernel

Similarity calculation is one of the most critical steps of clustering analysis, especially for arbitrarily formed elongated structures. When it comes to Density Peak Clustering (DPC), using Euclidean distance solely to calculate the similarity also makes it suffer arbitrarily formed data clustering. To tackle this deficiency of DPC, an improved Connectivity Kernel (ICK) was presented to accelerate Connectivity Kernel and help DPC identify clusters with arbitrarily formed structures, which mainly consist of two strategies:(i) Because that Connectivity Kernel suffers from outliers between two clusters if their density is as high as the backbone of the clusters, ICK firstly extracts local centers according to local density and relative location of points, which can eliminate most outliers and boundary points without breaking the original distribution of data. Thus, not only the adverse impact of outliers can be avoided, many meaningless calculations time can also be saved;(ii) ICK defines the connection between two local centers as their dissimilarity according to Connectivity Kernel. Differently, instead of traversing the entire dataset, ICK only focus on several specific path between two local centers to evaluate their connectivity, which further reduces the computational complexity of the algorithm to O(nlogn).Experiments on synthetic and real-world datasets demonstrate the effectiveness and robustness of the proposed algorithm in practical application.

A feasibility study for a unified, multimodal analysis of online information foraging in health-related topics.

Digital health literacy (DHL) is the ability to find, understand, and appraise online health-related information, as well as apply it to health behavior. It has become a core competence for navigating online information and health service environments. DHL involves solving ill-structured problems, where the problem and its solution are not clearcut and may have no single answer, such as in the process of sensemaking. We employ and expand on information foraging theory to address how experts and novices in information retrieval perform a search task. Our overarching aim is to pinpoint best practices and pitfalls in understanding and appraising health-related information online to develop a digital intervention to increase DHL and critical thinking. In this feasibility study, we recruited a total of twenty participants for our expert and novice subsamples. We collected sociodemographic data with a self-developed survey, video data through an observation protocol of a 10-minute search task, as well as audio-video data via a retrospective think-aloud. The three, multimodal data streams were transcribed and aligned. Codes were developed inductively in several iterations, then applied deductively to the entire dataset. Tabularized, coded and segmented qualitative data were used to create various quantitative models, which demonstrate viability for the qualitative and statistical comparison of our two subsamples. Data were visualized with Epistemic Network Analysis to analyze code co-occurrences in the three aligned data streams, and with Qualitative/Unified Exploration of State Transitions to examine the order in which participants in our two subsamples encountered online content. This paper describes our methods and planned analyses elaborated with mock figures. Quantifying qualitative data, aligning data streams, and representing all information in a tabularized dataset allows us to group data according to various participant attributes and employ data visualization techniques to pinpoint patterns therein.