Automatic Profiling Research Articles

Technologies of Data Protection and Institutional Decisions for Data Sovereignty

This paper aims to propose innovative actions of advanced technological solutions and consequent necessary institutional decisions to achieve in a reasonable time the definitive confidential data protection and data sovereignty, based on available scientific results. Confidential data protection is a fundamental and strategic issue in next-generation Internet systems to guarantee data sovereignty and the respect of human rights as stated in the foundation of the United Nations. Even if presently many international regulations are decisive steps to guarantee data protection within normative contexts, they are not adequate to face new technologies, such as facial recognition, automatic profiling, position tracking, biometric data, AI applications, and many others in the future, as they are implemented without any awareness by the interested subjects. Therefore, a new approach to data protection is mandatory based on innovative and disruptive technological solutions. A recent OECD report highlighted the need for the so-called Privacy-Enhancing Technologies (PETs) for the effective protection of confidential data, even more urgent for the coexistence of privacy and data sharing in international contexts. A common feature of these technologies is the use of software methodologies that can run on currently available microprocessors and their present immaturity. More effective and definitive protection can be achieved with another methodological approach based on the paradigm of ‘Data Usage Control’. This new concept guarantees data protection policy by default and initial design and it requires a new architecture of the data and a new HW&SW architecture of the computers. This contribution has a two-fold objective: first, to clarify why regulations alone and present technological proposals are not adequate for the effective and definitive protection of data and, second, to indicate the new necessary technological approach and the simultaneous institutional actions required to achieve the definitive protection and sovereignty of data in reasonable times, based on the results already available in the scientific literature.

Advancing preclinical drug evaluation through automated 3D imaging for high-throughput screening with kidney organoids

High-throughput drug screening is crucial for advancing healthcare through drug discovery. However, a significant limitation arises from available in vitro models using conventional 2D cell culture, which lack the proper phenotypes and architectures observed in three-dimensional (3D) tissues. Recent advancements in stem cell biology have facilitated the generation of organoids—3D tissue constructs that mimic human organs in vitro. Kidney organoids, derived from human pluripotent stem cells, represent a significant breakthrough in disease representation. They encompass major kidney cell types organized within distinct nephron segments, surrounded by stroma and endothelial cells. This tissue allows for the assessment of structural alterations such as nephron loss, a characteristic of chronic kidney disease. Despite these advantages, the complexity of 3D structures has hindered the use of organoids for large-scale drug screening, and the drug screening pipelines utilizing these complex in vitro models remain to be established for high-throughput screening. In this study, we address the technical limitations of kidney organoids through fully automated 3D imaging, aided by a machine-learning approach for automatic profiling of nephron segment-specific epithelial morphometry. Kidney organoids were exposed to the nephrotoxic agent cisplatin to model severe acute kidney injury. An U.S. Food and Drug Administration (FDA)-approved drug library was tested for therapeutic and nephrotoxicity screening. The fully automated pipeline of 3D image acquisition and analysis identified nephrotoxic or therapeutic drugs during cisplatin chemotherapy. The nephrotoxic potential of these drugs aligned with previous in vivo and human reports. Additionally, Imatinib, a tyrosine kinase inhibitor used in hematological malignancies, was identified as a potential preventive therapy for cisplatin-induced kidney injury. Our proof-of-concept report demonstrates that the automated screening process, using 3D morphometric assays with kidney organoids, enables high-throughput screening for nephrotoxicity and therapeutic assessment in 3D tissue constructs.

Discovering Functional Dependencies through Hitting Set Enumeration

Functional dependencies (FDs) are among the most important integrity constraints in databases. They serve to normalize datasets and thus resolve redundancies, they contribute to query optimization, and they are frequently used to guide data cleaning efforts. Because the FDs of a particular dataset are usually unknown, automatic profiling algorithms are needed to discover them. These algorithms have made considerable advances in the past few years, but they still require a significant amount of time and memory to process datasets of practically relevant sizes. We present FDHits, a novel FD discovery algorithm that finds all valid, minimal FDs in a given relational dataset. FDHits is based on several discovery optimizations that include a hybrid validation approach, effective hitting set enumeration techniques, one-pass candidate validations, and parallelization. Our experiments show that FDHits, even without parallel execution, has a median speedup of 8.1 compared to state-of-the-art FD discovery algorithms while using significantly less memory. This allows the discovery of all FDs even on datasets that could not be processed by the current state-of-the-art.

MesoGraph: Automatic profiling of mesothelioma subtypes from histological images

Mesothelioma is classified into three histological subtypes, epithelioid, sarcomatoid, and biphasic, according to the relative proportions of epithelioid and sarcomatoid tumor cells present. Current guidelines recommend that the sarcomatoid component of each mesothelioma is quantified, as a higher percentage of sarcomatoid pattern in biphasic mesothelioma shows poorer prognosis. In this work, we develop a dual-task graph neural network (GNN) architecture with ranking loss to learn a model capable of scoring regions of tissue down to cellular resolution. This allows quantitative profiling of a tumor sample according to the aggregate sarcomatoid association score. Tissue is represented by a cell graph with both cell-level morphological and regional features. We use an external multicentric test set from Mesobank, on which we demonstrate the predictive performance of our model. We additionally validate our model predictions through an analysis of the typical morphological features of cells according to their predicted score.

Automatic Profiles Collection from Twitter Users with Depressive Symptoms.

Mental illness is a pressing issue that needs urgent attention, as the number of people suffering from mental disorders continues to increase. Diagnosing mental health disorders can be challenging, and gathering information about a patient's medical history and symptoms is crucial for an accurate diagnosis. Self-disclosure on social media can provide valuable insights into whether users may be suffering from a mental illness. This paper proposes a method for automatically collecting data from social media users who disclosed their depression. The proposed approach yielded a 97% accuracy rate with a majority of 95%.

Working and working out: Decision-making inputs connect daily work demands to physical exercise.

Work demands can undermine engagement in physical exercise, posing a threat to employee health and well-being. Integrating resource theories and a novel decision-making theory called the decision triangle, we propose that this effect may emerge because work stress changes the energetic and emotional processes people engage in when making decisions about exercise after work. Using diary-style data across two workweeks (N = 83 workers, 783 days), we used multilevel latent profile analysis to extract common decision input profiles, or daily configurations of energy and affect as key decision-making resources. Consistent with the decision triangle, three profiles emerged: visceral inputs (low energy/high negative affect), automatic inputs (low energy/low negative affect), and logical inputs (high energy/low negative affect). Daily job demands were highest among the visceral profile. In turn, the daily visceral profile related to the lowest likelihood of and intensity of physical exercise after work, especially relative to the daily logical profile. Whether or not those in the daily automatic profile exercised depended on their health orientation, or trait-level value of maintaining personal health. Our results support decision-making as a promising mechanism explaining the link between work demands and healthy leisure choices. Organizational interventions can target work stress, health orientation, or logical decision-making to promote frequent and vigorous employee physical exercise. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

The Vision of the FAIR Digital Object Machine and Ubiquitous FDO Services

The Vision of the FAIR Digital Object Machine and Ubiquitous FDO Services

Autonomous Estimation of Patients’ Neuropsychological State Using Convolutional Neural Networks

The number of patients with neuropsychological problems is increasing rapidly in the world. Autonomous methods are replacing the traditional diagnosis methods in detection and classification of many mental and neurological problems. Machine learning algorithms and especially deep neural networks are able to diagnose various neurological and psychological complications automatically. In this paper, a machine learning based framework is used for autonomous estimation of patients’ neuropsychological state. The proposed framework can automatically diagnose neuropsychological state of the patients and present a personalized solution for their problems. A convolutional neural networks is used for automatic profiling of patients and to classify their mental state according to their EEG signals. The proposed framework can be used to help patients to have better life experience.

Text Mining for Employee Candidates Automatic Profiling Based on Application Documents

Opening job vacancies using the Internet will receive many applications quickly. Manually filtering resumes takes a lot of time and incurs huge costs. In addition, this manual screening process tends to be inaccurate due to fatigue conditions and fails in obtaining the right candidate for the job. This paper proposed a solution to automatically generate the most suitable candidate from the application document. In this study, 126 application documents from a private company were used for the experiment. The documents consist of 41 documents for Human Resource and Development (HRD) staff, 42 documents for IT (Data Developer), and 43 documents for the Marketing position. Text Processing is implemented to extract relevant information such as skills, education, experiences from the unstructured resumes and summarize each application. A specific dictionary for each vacancy is generated based on terms used in each profession. Two methods are implemented and compared to match and score the application document, namely Document Vector and N-gram analysis. The highest the score obtained by one document, the highest the possibility of application to be accepted. The two methods’ results are then validated by the real selection process by the company. The highest accuracy was achieved by the N-Gram method in IT vacancy with 87,5%, while the Document Vector showed 75% accuracy. For Marketing staff vacancy, both methods achieved the same accuracy as 78%. In HRD staff vacancy, the N-Gram method showed 68%, while Document Vector showed 74%. In conclusion, overall the N-gram method showed slightly better accuracy compared to the Document Vector method.

Application of Machine Learning Solutions to Optimize Parameter Prediction to Enhance Automatic NMR Metabolite Profiling.

The quality of automatic metabolite profiling in NMR datasets from complex matrices can be affected by the numerous sources of variability. These sources, as well as the presence of multiple low-intensity signals, cause uncertainty in the metabolite signal parameters. Lineshape fitting approaches often produce suboptimal resolutions to adapt them in a complex spectrum lineshape. As a result, the use of software tools for automatic profiling tends to be restricted to specific biological matrices and/or sample preparation protocols to obtain reliable results. However, the analysis and modelling of the signal parameters collected during initial iteration can be further optimized to reduce uncertainty by generating narrow and accurate predictions of the expected signal parameters. In this study, we show that, thanks to the predictions generated, better profiling quality indicators can be outputted, and the performance of automatic profiling can be maximized. Our proposed workflow can learn and model the sample properties; therefore, restrictions in the biological matrix, or sample preparation protocol, and limitations of lineshape fitting approaches can be overcome.

Research on Automatic Profile Generation Based on 3D Model

In this paper, the technology of profile generation based on 3D model is studied. The main steps are as follows: (1) the location where the profile needs to be generated in 3D model design; (2) Using 3D data cutting technology to realize the generation of geological lines in profile; (3) Read the basic exploration data related to profile position in the database; (4) According to the data generated in the first three steps, the cross-section is automatically drawn after data coordinate transformation. The above method can quickly generate the geological profile of any location according to the 3D geological model, which is helpful for geological analysis and provides reference data for engineering design.

Multilayer observation and estimation of the snowpack cold content in a humid boreal coniferous forest of eastern Canada

Abstract. Cold content (CC) is an internal energy state within a snowpack and is defined by the energy deficit required to attain isothermal snowmelt temperature (0 ∘C). Cold content for a given snowpack thus plays a critical role because it affects both the timing and the rate of snowmelt. Measuring cold content is a labour-intensive task as it requires extracting in situ snow temperature and density. Hence, few studies have focused on characterizing this snowpack variable. This study describes the multilayer cold content of a snowpack and its variability across four sites with contrasting canopy structures within a coniferous boreal forest in southern Québec, Canada, throughout winter 2017–2018. The analysis was divided into two steps. In the first step, the observed CC data from weekly snowpits for 60 % of the snow cover period were examined. During the second step, a reconstructed time series of modelled CC was produced and analyzed to highlight the high-resolution temporal variability of CC for the full snow cover period. To accomplish this, the Canadian Land Surface Scheme (CLASS; featuring a single-layer snow model) was first implemented to obtain simulations of the average snow density at each of the four sites. Next, an empirical procedure was used to produce realistic density profiles, which, when combined with in situ continuous snow temperature measurements from an automatic profiling station, provides a time series of CC estimates at half-hour intervals for the entire winter. At the four sites, snow persisted on the ground for 218 d, with melt events occurring on 42 of those days. Based on snowpit observations, the largest mean CC (−2.62 MJ m−2) was observed at the site with the thickest snow cover. The maximum difference in mean CC between the four study sites was −0.47 MJ m−2, representing a site-to-site variability of 20 %. Before analyzing the reconstructed CC time series, a comparison with snowpit data confirmed that CLASS yielded reasonable bulk estimates of snow water equivalent (SWE) (R2=0.64 and percent bias (Pbias) =-17.1 %), snow density (R2=0.71 and Pbias =1.6 %), and cold content (R2=0.93 and Pbias =-3.3 %). A snow density profile derived by utilizing an empirical formulation also provided reasonable estimates of layered cold content (R2=0.42 and Pbias =5.17 %). Thanks to these encouraging results, the reconstructed and continuous CC series could be analyzed at the four sites, revealing the impact of rain-on-snow and cold air pooling episodes on the variation of CC. The continuous multilayer cold content time series also provided us with information about the effect of stand structure, local topography, and meteorological conditions on cold content variability. Additionally, a weak relationship between canopy structure and CC was identified.

The selection of the main axis and the forming plane in automatic profiling and stamping lines

The article defines the main axis and the profiling plane in automatic profiling and stamping lines. Specific recommendations are given for choosing the position of the main axis and the profiling plane, depending on the configuration of the manufactured parts of the roll-formed section. Under the general name of profiling in the practice of stamping works, it is meant to obtain rigid and light profiles of large length and various configurations from sheet blanks. Profiling is carried out in four ways: in dies on crank presses, in dies on special bending presses, on universal bending machines (edging machines), on profiling roller machines. The first method, profiling on crank presses, is used for complex semi-closed and open profiles of relatively small length, if there are no special bending presses or profiling machines. The second method, profiling on special bending presses, is used for open and semiclosed profiles up to 5 mm long. The advantage of such presses is the possibility of using simple, and therefore cheap, tools in the manufacture. The third method, profiling on universal bending machines (edging machines), is used for bending parts (profiles of a simple shape in straight lines with different coupling radii determined by the radius of the machine ruler, for which the latter has a set of rulers). Bending machines allow bending materials of small thickness. Low productivity and the need for physical labor costs limit the use of these machines. The fourth method, profiling on roller machines, is used for open, semi-closed and closed profiles. The essence of the profiling process is to gradually change the profile drawing of a flat belt to a given profile when it is moved sequentially through several pairs of shaped rollers arranged sequentially one after the other in the same plane and rotating at the same speed. The article describes in detail the fourth method; the advantages and disadvantages are noted.

Self-Supervised Prototype Representation Learning for Event-Based Corporate Profiling

Event-based corporate profiling aims to assess the evolving operational status of the corresponding corporate from its event sequence. Existing studies on corporate profiling have partially addressed the problem via (i) case-by-case empirical analysis by leveraging traditional financial methods, or (ii) the automatic profile inference by reformulating the problem into a supervised learning task. However, both approaches heavily rely on domain knowledge and are labor-intensive. More importantly, the task-specific nature of both approaches prevents the obtained corporate profiles from being applied to diversified downstream applications. To this end, in this paper, we propose a Self-Supervised Prototype Representation Learning (SePaL) framework for dynamic corporate profiling. By exploiting the topological information of an event graph and exploring self-supervised learning techniques, SePaL can obtain unified corporate representations that are robust to event noises and can be easily fine-tuned to benefit various down-stream applications with only a few annotated data. Specifically, we first infer the initial cluster distribution of noise-resistant event prototypes based on latent representations of events. Then, we construct four permutation-invariant self-supervision signals to guide the representation learning of the event prototype. In terms of applications, we exploit the learned time-evolving corporate representations for both stock price spike prediction and corporate default risk evaluation. Experimental results on two real-world corporate event datasets demonstrate the effectiveness of SePaL for these two applications.

Automatic profile tracking of deformed surface in mirror milling based on ultrasonic measurement

Mirror milling is an effective and environmentally friendly means to manufacture large thin-walled parts. The remaining wall thickness of such parts is supposed to be strictly controlled to balance strength and weight reduction. However, the clamping deformation will lead to overcut or undercut if the nominal tool path is directly utilized. To address this issue, it would be helpful to scan and reconstruct the deformed unknown workpiece surface and then adjust the nominal tool path. In this paper, an automatic profile tracking approach is proposed to adaptively sample the deformed surface and reconstruct the surface model, based on a multi-ultrasonic-probe measurement system. The adaptive sampling algorithm calculates the position and surface normal of the next sampling point simultaneously based on a curvature sphere, which utilizes the surface information of the local area covered by the multi-ultrasonic-probe system. Moreover, a boundary processing algorithm is proposed to ensure the sampling process is carried out inside the boundary. The feasibility of the proposed automatic measurement method was validated through experiment.

A Comprehensive Model for the Use of Mechanisms for Automating Customs Operations

The article is devoted to the study of an important scientific and practical problem - the automation of customs operations. The purpose of the study was to develop an approach to the use of mechanisms for automating customs operations, including promising means of the Unified automated information system of customs authorities. The development of such an approach is based on the results of a study of the evolution of the system of customs operations in the Russian Federation, generalization of the problems of the formation of electronic documents and electronic information interaction of participants in customs operations, format-logical control of customs documents, the formation and application of automatic risk profiles, the implementation of expert checks during customs control. As a result of the study, modern principles of the functioning of information systems in customs were formulated, which made it possible to develop requirements for promising means of the Unified automated information system of customs authorities as an integrated mechanism for automating customs operations. On this basis, a comprehensive model for the use of mechanisms for automating customs operations has been proposed. As part of a comprehensive model, mechanisms for automating customs operations are considered, recommendations for their implementation are given, and possible effects from the use of these mechanisms are identified.

GroupTracer: Automatic Attacker TTP Profile Extraction and Group Cluster in Internet of Things

As Advanced Persistent Threat (APT) becomes increasingly frequent around the world, security experts are starting to look at how to observe, predict, and mitigate the damage from APT attacks. In the meantime, the Internet of things devices are also risky and heavily exposed to the Internet, making them more easily used by hacker organizations to launch APT attacks. An excellent attacker can take down millions of Internet of things devices in a short time. Once the IoT botnet is built, attackers can use it to launch complex attacks which could damage Internet infrastructure and cause network disconnection. This paper proposes GroupTracer, a framework for observing and predicting the Internet of things attacks. GroupTracer is designed to automatically extract the TTP profiles (i.e., tactics, techniques, and procedures) that can describe the behavior of attackers through their tactics, techniques, and processes and dig out the potential attacker groups behind complex attacks. Firstly, it captures attacks by IoT honeypots and extracts relevant fields from logs. Then, attack behaviors are automatically mapped to the ATT&CK framework to achieve automatic TTP profiles extraction. After that, GroupTracer presents four feature groups, including TTP profiles, Time, IP, and URL features, a total of 18 features, mines potential attack groups through hierarchical clustering algorithm, and compares the clustering results with two baseline algorithms. As the ground truth labels are unknown, we apply three internal validation indexes to evaluate the cluster quantity. Experimental results showed that the proposed framework has achieved an excellent performance in exploiting potential groups as the Calinski–Harabasz index reaches 3416.93. Eventually, attack trees are generated for each cluster where nodes indicate attack commands and edges represent command sequences. These attack trees could help better understand each attack group’s actions and techniques.

Attributed Collaboration Network Embedding for Academic Relationship Mining

Finding both efficient and effective quantitative representations for scholars in scientific digital libraries has been a focal point of research. The unprecedented amounts of scholarly datasets, combined with contemporary machine learning and big data techniques, have enabled intelligent and automatic profiling of scholars from this vast and ever-increasing pool of scholarly data. Meanwhile, recent advance in network embedding techniques enables us to mitigate the challenges of large scale and sparsity of academic collaboration networks. In real-world academic social networks, scholars are accompanied with various attributes or features, such as co-authorship and publication records, which result in attributed collaboration networks. It has been observed that both network topology and scholar attributes are important in academic relationship mining. However, previous studies mainly focus on network topology, whereas scholar attributes are overlooked. Moreover, the influence of different scholar attributes are unclear. To bridge this gap, in this work, we present a novel framework of Attributed Collaboration Network Embedding (ACNE) for academic relationship mining. ACNE extracts four types of scholar attributes based on the proposed scholar profiling model, including demographics, research, influence, and sociability. ACNE can learn a low-dimensional representation of scholars considering both scholar attributes and network topology simultaneously. We demonstrate the effectiveness and potentials of ACNE in academic relationship mining by performing collaborator recommendation on two real-world datasets and the contribution and importance of each scholar attribute on scientific collaborator recommendation is investigated. Our work may shed light on academic relationship mining by taking advantage of attributed collaboration network embedding.

Research on Automatic Control Mode of Profile Cutting Machine Based on PLC

Aiming at the constant need of cutting various metal profiles in industry, the control mode of automatic profile cutting machine is studied, which is based on PLC (Programmable Logic Controller) controller. MCGS (Monitor and Control Generated System) is used for monitoring and management, and the modular design of control layer is completed. The core of the control layer is that we have a cutting material database, which makes up for the fact that the current cutting machine control system only aims at a single cutting model or material, and the cutting speed and torque of the cutting machine need to be adjusted or designed manually. Based on this market demand, the automatic control mode of profile cutting machine based on PLC is established. Practice shows that this mode is based on the analysis and comparison of common cutting materials in the market, and the material database is consequently established. Users can choose the required cutting mode through the intuitive MCGS graphical interface, which improves the cutting efficiency of users and reduces the development cost of enterprises.

BASICS OF AUTOMATION OF PROFILING OF HIGH-EFFICIENT INSERTS OF THE TOOL FOR THREAD MACHINING

Many modern threaded joints especially in the oil and gas industry, must satisfy high requirements for steel mechanical properties of which they are made. To ensure efficient threading of such hard-to-machine materials (ultimate strength is more than 1250 MPa), it is necessary to use carbide cutting inserts with the negative rake angle at its corner. However, modern manufacturers of such inserts make them only with a zero value of the rake angle. This is due to the difficulty of profiling of the insert cutting edge. This article proposes the basics of such inserts automatic profiling which is based on the algorithm developed by the authors. It includes two corrective calculations of the insert cutting edge profile, namely: the radial displacement of the corner and the tangential displacement of the area forming the crest of the thread. As a result, the computer application with the input parameter of the rake angle and output parameter of the upgraded profile of the insert cutting edge is received.