Interplay of Machine Learning and Software Engineering for Quality Estimations

Abubakar, Hamza; Obaidat, M. S.; Gupta, Aaryan; Bhattacharya, Pronaya; Tanwar, Sudeep

doi:10.1109/ccci49893.2020.9256507

Cited by 4 publications

(1 citation statement)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Classification: For this phase, Most researchers employed machine learning techniques or pattern recognition 77 . Under different situations, such as pedestrian walking pace (slow, normal, fast), apparel (sports, jeans), and body measurements (weight, height), the classifier is designed to differentiate between four typical smartphone carrying ways: (1) swinging, (2) texting, (3) talking, and (4) pocket.…”

Section: Stride Length Calculation Modelsmentioning

confidence: 99%

A survey of step length estimation models based on inertial sensors for indoor navigation systems

Soni

Trapasiya²

2021

Int J Communication

View full text Add to dashboard Cite

Summary In many applications (i.e., navigation of vehicles in traffic, museum guide, guidance to firefighters in poor visibility, and elderly care assistance systems), location‐based services (LBS) rely on the global positioning system (GPS) for accurate positioning. LBS can be challenging to employ in an indoor environment because of the lack of a line‐of‐sight (LOS) component and other signal propagation issues. Over the last decade, researchers worldwide have been motivated to solve the problem of continuous indoor positioning in situations when GPS is ineffective. We thoroughly reviewed research conducted by many scholars and compiled this survey article, which provides a comprehensive overview of indoor pedestrian navigation systems, prevalent step length estimation models, and the current state of the art. There are two key strategies in which substantial work on indoor localization is carried out. One is a dead reckoning in which the present location of the pedestrian is calculated based on details about previous location, and the other is an inertial navigation system in which the forward acceleration from an inertial sensor strapped down to the user's body is integrated twice to establish the current location. The step length is a crucial component of the inertial sensor‐based navigation in both these strategies. For different users, step length is variable and also changes at different walking speeds for each user. As a result, the focus of this survey is on step length estimation models. The survey describes and compares various approaches to estimating step length from various perspectives, including the research method used, the length of the test path, various walking speeds, the location of the sensor device on the user's body, and the accuracy achieved in estimating step length. Because inertial sensors are embedded in smartphones, it is vital to identify the smartphone mode (i.e., phoning, texting, or pocket) during each step taken while using smartphone‐based indoor navigation. The report concludes with a quantitative evaluation of the performance of smartphone mode recognition modules. At the end, we list out challenges still to be addressed for scholars working on indoor navigation.

show abstract

Section: Stride Length Calculation Modelsmentioning

confidence: 99%

A survey of step length estimation models based on inertial sensors for indoor navigation systems

Soni

Trapasiya²

2021

Int J Communication

View full text Add to dashboard Cite

show abstract

The Impact of Resource Allocation on the Machine Learning Lifecycle

Duda,

Hofmann,

Urbach

et al. 2023

Bus Inf Syst Eng

View full text Add to dashboard Cite

An organization’s ability to develop Machine Learning (ML) applications depends on its available resource base. Without awareness and understanding of all relevant resources as well as their impact on the ML lifecycle, we risk inefficient allocations as well as missing monopolization tendencies. To counteract these risks, the study develops a framework that interweaves the relevant resources with the procedural and technical dependencies within the ML lifecycle. To rigorously develop and evaluate this framework the paper follows the Design Science Research paradigm and builds on a literature review and an interview study. In doing so, it bridges the gap between the software engineering and management perspective to advance the ML management discourse. The results extend the literature by introducing not yet discussed but relevant resources, describing six direct and indirect effects of resources on the ML lifecycle, and revealing the resources’ contextual properties. Furthermore, the framework is useful in practice to support organizational decision-making and contextualize monopolization tendencies.

show abstract