Data-driven Systems Research Articles

Monitoring of Laser Powder Bed Fusion process by bridging dissimilar process maps using deep learning-based domain adaptation on acoustic emissions

Advances in sensorization and identification of information embedded inside sensor signatures during manufacturing processes using Machine Learning (ML) algorithms for better decision-making have become critical enablers in building data-driven monitoring systems. In the Laser Powder Bed Fusion (LPBF) process, data-driven-based process monitoring is gaining popularity since it allows for real-time component quality verification. Real-time qualification of the additively manufactured parts has a significant advantage as the cost of conventional post-manufacturing inspection methods can be reduced. Also, corrective actions or build termination could be done to save machine time and resources. However, despite the successful development in addressing monitoring needs in LPBF processes, less research has been paid to the ML model's robustness in decision-making when dealing with variations in data distribution from the laser-material interaction owing to different process spaces. Inspired by the idea of domain adaptation in ML, in this work, we propose a deep learning-based unsupervised domain adaptation technique to tackle shifts in data distribution owing to different process parameter spaces. The temporal waveforms of acoustic emissions from the LPBF process zone corresponding to three regimes, namely Lack of Fusion, conduction, and keyhole, were acquired on two different 316 L stainless steel powder distributions (> 45 µm and < 45 µm) with two different parameter sets. Temporal and spectral analysis of the acoustic waveforms corresponding to the powder distributions treated with different laser parameters showed the presence of offset in the data distribution, which was subsequently treated with the proposed unsupervised domain adaptation technique to have an ML model that could be generalized. Furthermore, the prediction accuracy of the proposed methodology between the two distributions showed the feasibility of adapting to the newer environment unsupervisedly and improving the ML model's generalizability.

Data-Driven Decision Support Systems for Business Process Improvement

The accessibility of data is altering how businesses make decisions at different levels. Scholars and professionals are investigating the ways in which Business Process suppliers can profit from the availability and application of data, particularly in relation to decision-making concerning service provision. Business Process Improvement is one of the applications that is anticipated to gain the most from the accessibility of information. Suppliers of services can avoid failures by making prompt and well-informed decisions based on the evaluation of the resource's health state. Despite this, providing data-driven BPI services is not simple, and providers must set up their systems to correctly gather, process, and utilize past and current data. This study introduces a data-driven business intelligence framework to provide use full insights for improving business process activities. This framework offers a set of visualization tools that help interpret the relation between different factors that can improve the management of different business processes. Moreover, our framework provides successful integration of random forests to allow predictive modeling of sales, profits, and discounts across different regions.

Advancing Sensor Data Integrity with Deep Learning-Based Fault Detection

In the realm of modern data-driven systems, accurate and reliable sensor measurements are imperative for informed decision-making and system integrity. The objective of this project is to develop a robust sensor fault detection methodology leveraging the power of deep learning techniques. The ubiquity of sensors such as temperature and humidity sensors has led to a critical need for discerning between accurate readings and erroneous data, thereby enhancing the reliability of these measurements. This study addresses the prevailing challenge of sensor reliability by introducing a data-driven approach that harnesses deep learning algorithms to detect sensor faults promptly and accurately. A comprehensive dataset comprising sensor readings under diverse conditions forms the foundation of this investigation. Multiple cutting-edge deep learning architectures and techniques are systematically explored and evaluated against this dataset to identify the most efficient and precise sensor fault detection method. The outcomes of this research contribute significantly to advancing the state-of-the-art in sensor fault detection, with implications for a wide array of applications reliant on sensor measurements. By harnessing the capabilities of deep learning, this project presents a tangible solution to the challenge of accurately identifying faulty sensor data in real-time, thereby bolstering the dependability and efficacy of sensor-driven systems. Ultimately, this work underscores the potential of integrating advanced machine learning techniques in ensuring the reliability and precision of sensor data, heralding a new era of robust and trustworthy sensor-enabled environments.

Data Science Principles for Interpretable and Explainable AI

Society’s capacity for algorithmic problem-solving has never been greater. Artificial Intelligence is now applied across more domains than ever, a consequence of powerful abstractions, abundant data, and accessible software. As capabilities have expanded, so have risks, with models often deployed without fully understanding their potential impacts. Interpretable and interactive machine learning aims to make complex models more transparent and controllable, enhancing user agency. This review synthesizes key principles from the growing literature in this field. We first introduce precise vocabulary for discussing interpretability, like the distinction between glass box and explainable models. We then explore connections to classical statistical and design principles, like parsimony and the gulfs of interaction. Basic explainability techniques – including learned embeddings, integrated gradients, and concept bottlenecks – are illustrated with a simple case study. We also review criteria for objectively evaluating interpretability approaches. Throughout, we underscore the importance of considering audience goals when designing interactive data-driven systems. Finally, we outline open challenges and discuss the potential role of data science in addressing them. Code to reproduce all examples can be found at https://go.wisc.edu/3k1ewe.

DATA REFUSAL FROM BELOW: A FRAMEWORK FOR UNDERSTANDING, EVALUATING, AND ENVISIONING REFUSAL STRATEGIES

Amidst calls for public accountability over large data-driven systems, feminist and indigenous scholars have developed refusal as a practice that challenges the authority of data collectors. However, because data affects so many aspects of daily life, it can be hard to see seemingly different refusal strategies as part of the same repertoire. Furthermore, conversations about refusal often happen from the standpoint of designers and policymakers rather than the people and communities most affected by data collection. In this paper, we introduce a framework for data refusal from below—writing from the standpoint of people who refuse, rather than the institutions that seek their compliance. We characterize refusal strategies across four constituent facets common to all refusal, whatever tactics are used: autonomy, or how refusal accounts for individual and collective interests; time, or whether refusal reacts to past harm or proactively prevents future harm; power, or the extent to which refusal makes change possible; and cost, or whether or not refusal can reduce or redistribute penalties experienced by refusers. We illustrate each facet by drawing on cases of people and collectives that have refused data systems. Together, the four facets of our framework are designed to help scholars and activists describe, evaluate, and imagine new forms of refusal.

Evaluating the performance of machine learning CFD-based and hybrid analytical models for transient flow prediction in temperature-compensated digital flow units

This investigation utilized binary-coded, parallel-connected on-off valves that can achieve high flow rates with fewer valves while addressing flow peak challenges. By considering temperature and refining modeling techniques, the study rectifies certain limitations observed in previous research, such as neglecting temperature, imprecise valve orifice flow coefficients, and absent flow pattern visualization, thereby enhancing flow prediction accuracy.The results for the ML_CFD-based model suggest that although extrapolation challenges exist in rarely data-driven systems, the proposed approach exhibits errors under 5 % across diverse metrics, attributable to the effectiveness of well-constrained overparameterized models and the segmented structure of digital flow control units. On the other hand, while the simplified Hybrid Analytical model shows minor deviations and offers easier implementation, it encounters constraints when processing data beyond its pre-tuned coefficient of discharge values.

Two-stage video-based convolutional neural networks for adult spinal deformity classification.

Assessment of human gait posture can be clinically effective in diagnosing human gait deformities early in life. Currently, two methods-static and dynamic-are used to diagnose adult spinal deformity (ASD) and other spinal disorders. Full-spine lateral standing radiographs are used in the standard static method. However, this is a static assessment of joints in the standing position and does not include information on joint changes when the patient walks. Careful observation of long-distance walking can provide a dynamic assessment that reveals an uncompensated posture; however, this increases the workload of medical practitioners. A three-dimensional (3D) motion system is proposed for the dynamic method. Although the motion system successfully detected dynamic posture changes, access to the facilities was limited. Therefore, a diagnostic approach that is facility-independent, has low practice flow, and does not involve patient contact is required. We focused on a video-based method to classify patients with spinal disorders either as ASD, or other forms of ASD. To achieve this goal, we present a video-based two-stage machine-learning method. In the first stage, deep learning methods are used to locate the patient and extract the area where the patient is located. In the second stage, a 3D CNN (convolutional neural network) device is used to capture spatial and temporal information (dynamic motion) from the extracted frames. Disease classification is performed by discerning posture and gait from the extracted frames. Model performance was assessed using the mean accuracy, F1 score, and area under the receiver operating characteristic curve (AUROC), with five-fold cross-validation. We also compared the final results with professional observations. Our experiments were conducted using a gait video dataset comprising 81 patients. The experimental results indicated that our method is effective for classifying ASD and other spinal disorders. The proposed method achieved a mean accuracy of 0.7553, an F1 score of 0.7063, and an AUROC score of 0.7864. Additionally, ablation experiments indicated the importance of the first stage (detection stage) and transfer learning of our proposed method. The observations from the two doctors were compared using the proposed method. The mean accuracies observed by the two doctors were 0.4815 and 0.5247, with AUROC scores of 0.5185 and 0.5463, respectively. We proved that the proposed method can achieve accurate and reliable medical testing results compared with doctors' observations using videos of 1 s duration. All our code, models, and results are available at https://github.com/ChenKaiXuSan/Walk_Video_PyTorch. The proposed framework provides a potential video-based method for improving the clinical diagnosis for ASD and non-ASD. This framework might, in turn, benefit both patients and clinicians to treat the disease quickly and directly and further reduce facility dependency and data-driven systems.

Adversarial Security Verification of Data-Driven FDC Systems

Adversarial Security Verification of Data-Driven FDC Systems

Writing in the Air: Unconstrained Text Recognition From Finger Movement Using Spatio-Temporal Convolution

We introduce a new benchmark dataset for the challenging writing in the air (WiTA) task—an elaborate task bridging vision and NLP. WiTA implements an intuitive and natural writing method with finger movement for HCI. Our WiTA dataset will facilitate the development of data-driven WiTA systems which thus far have displayed unsatisfactory performance—due to lack of dataset as well as traditional statistical models they have adopted. Our dataset consists of five sub-datasets in two languages (Korean and English) and amounts to 209,926 video instances from 122 participants. We capture finger movement for WiTA with RGB cameras to ensure wide accessibility and cost-efficiency. Next, we propose spatio-temporal residual network architectures inspired by 3D ResNet. These models perform unconstrained text recognition from finger movement, guarantee a real-time operation ( <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$&gt;$</tex-math></inline-formula> 100 FPS), and will serve as an evaluation standard. Our dataset and the source codes are available at <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/Uehwan/WiTA</uri> .

Preface

The 2023 3rd International Conference on Electrical Engineering and Computer Technology (ICEECT 2023), a virtual conference, was held during September 22nd–24th, 2023 in Nanjing, China, bringing together about 80 participants around the world.It is well known that technological advancements have made our learning continuous. It is high time for engineers and technocrats to think about innovation and contribute to technological advancements. Technological advancements at an exceptional rate could only take place by exchanging thoughts, ideas, innovations, problems, real-time solutions to the problems and technologies.ICEECT 2023 has provided such a platform for researchers, engineers, academicians, industrial professionals, and practitioners of Electrical Engineering and Computer Technology fields from all over the globe. It has kindled validated discussions on the technological advancements, recent trends, various issues of sustainability, and the solutions adopted to address these issues in the amalgamating fields of Electrical Engineering and Computer Technology by embodying innovative ideas and pivoting advanced technologies. Emphasis was attributed to the recent developments in the fields of Electrical Engineering and Computer Technology.The Conference was made stimulating and informative by keynote speeches delivered by the pioneers of the fields from different corners of the world as well as many oral and poster presentations. Among the many speakers, Marco C. Campi, acting as one of our keynote speakers, is a Professor of Automatic Control at the University of Brescia, Italy, where he has taught topics related to data-driven control, systems theory, and inductive methods for more than twenty-five years. He has delivered plenary addresses at major conferences, including CDC, MTNS, and SYSID. And another keynote speaker Lei Chen is currently an Associate Professor at the School of Information Science and Engineering, Shandong University, China. His research interests include Image Processing and Computer Vision, Visual Quality Assessment and Pattern Recognition, Machine Learning and Artificial Intelligence. They both made dramatic speeches employing their research experience and professional insights. What’s more, attendees interacted actively with each other, exchanging views on the topics concerned and sharing their unique experience on the development trends and research directions.We would like to address our sincere thanks to the eminent speakers, the Technical Committee members, authors, reviewers, participants, the esteemed associates of the Conference, and all the members who have supported the Conference. Special acknowledgement to the staff of Journal of Physics: Conference Series for all their efforts in making this volume published.The Committee of ICEECT 2023List of Committee Member are availble in this pdf.

A Core Outcome Set for Inflammatory Bowel Diseases: Development and Recommendations for Implementation in Clinical Practice Through an International Multi-stakeholder Consensus Process.

Standardising health outcome measurements supports delivery of care and enables data-driven learning systems and secondary data use for research. As part of the Health Outcomes Observatory [H2O] initiative, and building on existing knowledge, a core outcome set [COS] for inflammatory bowel diseases [IBD] was defined through an international modified Delphi method. Stakeholders rated 90 variables on a 9-point importance scale twice, allowing score modification based on feedback displayed per stakeholder group. Two consecutive consensus meetings were held to discuss results and formulate recommendations for measurement in clinical practice. Variables scoring 7 or higher by ≥80% of the participants, or based on consensus meeting agreement, were included in the final set. In total, 136 stakeholders (45 IBD patients [advocates], 74 health care professionals/researchers, 13 industry representatives, and four regulators) from 20 different countries participated. The final set includes 18 case-mix variables, three biomarkers [haemoglobin to detect anaemia, C-reactive protein and faecal calprotectin to detect inflammation] for completeness, and 28 outcomes (including 16 patient-reported outcomes [PROs] and one patient-reported experience). The PRO-2 and IBD-Control questionnaires were recommended to collect disease-specific PROs at every contact with an IBD practitioner, and the Subjective Health Experience model questionnaire, PROMIS Global Health and Self-Efficacy short form, to collect generic PROs annually. A COS for IBD, including a recommendation for use in clinical practice, was defined. Implementation of this set will start in Vienna, Berlin, Barcelona, Leuven, and Rotterdam, empowering patients to better manage their care. Additional centres will follow worldwide.

A theme section on the central role of modeling in designing and explaining data-driven systems and software

A theme section on the central role of modeling in designing and explaining data-driven systems and software

Universal adversarial backdoor attacks to fool vertical federated learning

Vertical federated learning (VFL) is a privacy-preserving distribution learning paradigm that enables participants, owning different features of the same sample space to train a machine learning model collaboratively while retaining their data locally. This paradigm facilitates improved efficiency and security for participants such as financial or medical fields, making VFL an essential component of data-driven Artificial Intelligence systems. Nevertheless, the partitioned structure of VFL can be exploited by adversaries to inject a backdoor, enabling them to manipulate the VFL predictions. In this paper, we aim to investigate the vulnerability of VFL in the context of binary classification tasks. To this end, we define a threat model for backdoor attacks in VFL and introduce a universal adversarial backdoor (UAB) attack to poison the predictions of VFL. The UAB attack, consisting of universal trigger generation and clean-label backdoor injection, is incorporated during the VFL training at specific iterations. This is achieved by alternately optimizing VFL sub-problems' universal trigger and model parameters. Our work distinguishes itself from existing studies on designing backdoor attacks for VFL, as those require the knowledge of auxiliary information that is not accessible within the split VFL architecture. In contrast, our approach does not require additional data to execute the attack. On the real-world datasets, our approach surpasses existing state-of-the-art methods, achieving up to 100% backdoor task performance while maintaining the main task performance. Our results in this paper make a major advance in revealing the hidden backdoor risks of VFL, hence paving the way for the future development of secure VFL. Our results in this paper make a major advance in revealing the hidden backdoor risks of VFL, hence paving the way for the future development of secure VFL applications such as finance.

ESFILES: Intelligent process flowsheet synthesis using process knowledge, symbolic AI, and machine learning

Process flowsheet synthesis, design, and simulation require integrated approaches that combine domain knowledge and data-driven methods for fast, efficient, and reliable solutions. However, due to the recent surge in data and machine learning capabilities, there has been a shift towards building purely data-driven systems for process flowsheet synthesis and related problems. Such approaches have certain drawbacks. Here, we present a hybrid method that combines data-driven approaches with domain knowledge to represent process flowsheets and solve problems related to process synthesis, design, and simulation. We present an extended SFILES (or eSFILES) representation, a multi-level hierarchical flowsheet representation with varying degrees of process knowledge. At level 0, flow diagrams are represented as purely text-based SFILES strings. At level 1, the SFILES grammar, along with inferencing algorithms, is used to construct a flowsheet hypergraph explicitly representing flow diagram connectivity. At level 2, specifications needed for material and energy balance calculations are introduced, and, after simulation, the results are also added using annotated flowsheet hypergraphs. Finally, at level 3, a process ontology is connected with the annotated flowsheet hypergraph to include design and operation parameters as well as the detailed simulation results. We discuss this hierarchical framework using several case studies.

Dual-Signal Feature Spaces Map Protein Subcellular Locations Based on Immunohistochemistry Image and Protein Sequence.

Protein is one of the primary biochemical macromolecular regulators in the compartmental cellular structure, and the subcellular locations of proteins can therefore provide information on the function of subcellular structures and physiological environments. Recently, data-driven systems have been developed to predict the subcellular location of proteins based on protein sequence, immunohistochemistry (IHC) images, or immunofluorescence (IF) images. However, the research on the fusion of multiple protein signals has received little attention. In this study, we developed a dual-signal computational protocol by incorporating IHC images into protein sequences to learn protein subcellular localization. Three major steps can be summarized as follows in this protocol: first, a benchmark database that includes 281 proteins sorted out from 4722 proteins of the Human Protein Atlas (HPA) and Swiss-Prot database, which is involved in the endoplasmic reticulum (ER), Golgi apparatus, cytosol, and nucleoplasm; second, discriminative feature operators were first employed to quantitate protein image-sequence samples that include IHC images and protein sequence; finally, the feature subspace of different protein signals is absorbed to construct multiple sub-classifiers via dimensionality reduction and binary relevance (BR), and multiple confidence derived from multiple sub-classifiers is adopted to decide subcellular location by the centralized voting mechanism at the decision layer. The experimental results indicated that the dual-signal model embedded IHC images and protein sequences outperformed the single-signal models with accuracy, precision, and recall of 75.41%, 80.38%, and 74.38%, respectively. It is enlightening for further research on protein subcellular location prediction under multi-signal fusion of protein.

Information Privacy Analysis: The USA Perspective

Abstract: Informational privacy has become a cornerstone of individual liberties in the United States, gaining prominence in an age where digital technologies and data-driven systems permeate every aspect of daily life. This paper explores the intricate fabric of the United States’ Constitutional Right to Informational Privacy, meticulously tracing its historical roots, dissecting pivotal legal precedents, scrutinizing legislative milestones, and shedding light on contemporary challenges. The paper commences by unraveling the historical foundations of informational privacy, harking back to early legal and philosophical tenets that laid the groundwork for the right to privacy as we know it today. It then pivots to a meticulous analysis of the Fourth Amendment, which, through a series of landmark Supreme Court decisions, extended protection against unreasonable searches and seizures to encompass the realm of electronic communications and data. In an era defined by rapid technological advancements, section four of the paper scrutinizes the implications of the digital age on privacy rights. From electronic surveillance to data collection and cybersecurity, this section elucidates the multifaceted landscape in which informational privacy operates. Key Supreme Court cases, such as Griswold v. Connecticut and Roe v. Wade, which have significantly expanded the ambit of informational privacy, are exhaustively examined in section five. The paper then transitions to section six to an exploration of legislative developments that have sought to safeguard informational privacy, including the Electronic Communications Privacy Act (ECPA) and other federal and state laws like HIPAA, COPPA, and CCPA. However, in an era rife with technological innovations and data-driven commerce, challenges and controversies abound. Section seven investigates ongoing debates surrounding informational privacy, including the conundrum of government surveillance, the specter of data breaches, and the intricacies of online privacy regulation. Section eight delves into the delicate balance between privacy and national security, particularly in the wake of post-9/11 policies and practices. The inherent tension between individual privacy rights and the imperatives of national security forms a critical facet of the contemporary discourse on informational privacy. Anticipating the ever-evolving landscape of information technology and data governance, section nine explores emerging trends and potential future directions in informational privacy law. This section scrutinizes proposed legislation and anticipates the impact of technological advancements on the evolution of privacy rights. Finally, this paper synthesizes the historical trajectory and current state of the constitutional right to informational privacy in the United States. It underscores the enduring relevance of this right in a society where data has become both a precious commodity and a potential threat. As information technology continues to redefine the boundaries of privacy, the protection of informational privacy remains an essential facet of individual liberty and autonomy

Making recommender systems forget: Learning and unlearning for erasable recommendation

Regulations now mandate data-driven systems, e.g., recommender systems, to empower users to delete private individual data. This prompts the crucial unlearning of data from machine learning models, spotlighting the understudied machine unlearning problem. Despite the widespread usage of machine learning models in modern-day recommender systems, unlearning in this context lacks attention. Existing unlearning methods fall short in the preservation of collaborative information across users and items. To bridge this gap, we propose LASER: a model-agnostic erasable recommendation framework. LASER partitions data into disjoint and balanced shards using hypergraph-based embeddings. Sequential training on these shards, augmented by curriculum learning, LASER sufficiently preserves collaborative information and refines model utility. To address the inefficiency of sequential training, we integrate early stopping and parameter manipulation. Our theoretical analyses and real-world dataset experiments validate LASER’s effectiveness. It enables efficient unlearning while outperforming state-of-the-art models in preserving model utility.

Lessons Learned in Transcribing 5000 h of Air Traffic Control Communications for Robust Automatic Speech Understanding

Voice communication between air traffic controllers (ATCos) and pilots is critical for ensuring safe and efficient air traffic control (ATC). The handling of these voice communications requires high levels of awareness from ATCos and can be tedious and error-prone. Recent attempts aim at integrating artificial intelligence (AI) into ATC communications in order to lessen ATCos’s workload. However, the development of data-driven AI systems for understanding of spoken ATC communications demands large-scale annotated datasets, which are currently lacking in the field. This paper explores the lessons learned from the ATCO2 project, which aimed to develop an unique platform to collect, preprocess, and transcribe large amounts of ATC audio data from airspace in real time. This paper reviews (i) robust automatic speech recognition (ASR), (ii) natural language processing, (iii) English language identification, and (iv) contextual ASR biasing with surveillance data. The pipeline developed during the ATCO2 project, along with the open-sourcing of its data, encourages research in the ATC field, while the full corpus can be purchased through ELDA. ATCO2 corpora is suitable for developing ASR systems when little or near to no ATC audio transcribed data are available. For instance, the proposed ASR system trained with ATCO2 reaches as low as 17.9% WER on public ATC datasets which is 6.6% absolute WER better than with “out-of-domain” but gold transcriptions. Finally, the release of 5000 h of ASR transcribed speech—covering more than 10 airports worldwide—is a step forward towards more robust automatic speech understanding systems for ATC communications.

Open space dynamics during coronavirus disease-2019

Although numerous studies have been carried out to examine the general trend of urban mobility during coronavirus disease-2019 (COVID-19), there is not enough research on changes in pedestrian behavioural characteristics and crowd dynamics in public spaces. Understanding and monitoring such changes are critical for the better management and design of public open space in case of future outbreaks of infectious diseases. To fill this gap, pedestrian movements are tracked and analysed with deep learning-based video analytics based on anonymised video footage along a major promenade in Hong Kong before and during COVID-19. Specifically, comparisons were made on pedestrian-flow characteristics, pedestrian activities and social distancing. Then, this study examines the dynamics of pedestrian crowding under different scenarios, using agent-based simulation. Model results suggest that the public space was characterised by fewer visitors, a higher average walking speed, a higher percentage of people exercising and a lower percentage of people carrying out stationary activities during COVID-19. In addition, a higher level of voluntary social distancing was observed. Several hot spots for pedestrian crowding were also identified. Learning from the above, it is suggested that multifunctional public space should be designed; and data-driven visitor management systems should be established to prepare for different scenarios in future cities.

Rethinking People Analytics With Inverse Transparency by Design

Employees work in increasingly digital environments that enable advanced analytics. Yet, they lack oversight over the systems that process their data. That means that potential analysis errors or hidden biases are hard to uncover. Recent data protection legislation tries to tackle these issues, but it is inadequate. It does not prevent data misusage while at the same time stifling sensible use cases for data. We think the conflict between data protection and increasingly data-driven systems should be solved differently. When access to an employees' data is given, all usages should be made transparent to them, according to the concept of inverse transparency. This allows individuals to benefit from sensible data usage while addressing the potentially harmful consequences of data misusage. To accomplish this, we propose a new design approach for workforce analytics software we refer to as inverse transparency by design. To understand the developer and user perspectives on the proposal, we conduct two exploratory studies with students. First, we let small teams of developers implement analytics tools with inverse transparency by design to uncover how they judge the approach and how it materializes in their developed tools. We find that architectural changes are made without inhibiting core functionality. The developers consider our approach valuable and technically feasible. Second, we conduct a user study over three months to let participants experience the provided inverse transparency and reflect on their experience. The study models a software development workplace where most work processes are already digital. Participants perceive the transparency as beneficial and feel empowered by it. They unanimously agree that it would be an improvement for the workplace. We conclude that inverse transparency by design is a promising approach to realize accepted and responsible people analytics.