End-user Interface Research Articles

Open source platform for Estonian speech transcription

AbstractThis paper presents our progress in developing and maintaining a public speech and speaker recognition platform for the Estonian language. The platform consists of a speech processing pipeline and a web-based user interface for end-users, offering transcript post-editing functionality. It is offered for free as a public service and is in active use. The service provides significantly higher speech recognition accuracy than commercial alternatives. We discuss the switch to a workflow management system and how it has improved the core speech processing pipeline. The core systems behind the platform have been made available as open-source code and deployed internally by multiple public and private institutions.

Implementasi Framework Streamlit Sebagai Prediksi Harga Jual Rumah Dengan Linear Regresi

This research aims to develop an Artificial Intelligence (AI)-based application using the Streamlit framework to predict house sale prices in Banjarmasin City using Linear Regression methodology. The increase in demand and supply of properties in Banjarmasin City poses a complex challenge in determining house sale prices. The Linear Regression method was chosen as the primary analytical tool to identify factors influencing house sale prices. This application utilizes historical data of house sale prices and variables such as land area, building area, number of rooms, proximity to public facilities, and geographical location as inputs for the Linear Regression model. Furthermore, the Streamlit framework is employed to create an interactive and user-friendly interface for end-users. The outcome of this research is an AI application that assists potential buyers or sellers in Banjarmasin City in determining competitive prices. By inputting information about the property being evaluated, users can obtain a more accurate estimated sale price based on factors identified by the Linear Regression model. In testing the application, actual house sale price data from Banjarmasin City was used to assess the model's accuracy. The testing results indicate that the application is capable of providing reasonably accurate price estimates, achieving an accuracy level of 67.8%. Thus, this AI application holds the potential to be a valuable tool in the property industry in Banjarmasin City, aiding stakeholders in more informed and data-driven decision-making regarding house sale prices. Additionally, this application could serve as a foundation for further developments in AI research and property price analysis.

Spinach fungi guard: A deep learning-based software solution for swift detection and remediation of fungal diseases in spinach leaves

We have developed a software solution aimed at assisting farmers in quickly detecting and finding a cure for fungal diseases in spinach leaves. This solution utilizes Deep Learning techniques, specifically a Convolutional Neural Network (CNN), to effectively detect diseases caused by fungi in spinach leaves and suggest suitable remedies to the user. To train our model; we gathered a spinach fungi disease image dataset consisting of over 700 images categorized into six classes, including five diseases and one healthy category. Prior to model training, we resized the images and applied various image augmentation techniques to improve the robustness of the model. Using Keras, we constructed a sequential CNN model for disease classification. The model was trained on the dataset and evaluated on the validation set, achieving an impressive accuracy of 89.86 %. To provide an intuitive interface for end users, we implemented a PySide2-based GUI application that leverages the trained model to classify disease in spinach leaf images provided as input.Our software not only accurately classifies the disease but also suggests appropriate remedies or medications for the specific disease. Furthermore, it provides links to relevant products on various e-commerce sites, enabling users to conveniently purchase the required medications. This comprehensive solution empowers end users to analyze infected spinach leaf images, accurately classify diseases, and take necessary actions by applying appropriate remedies and acquiring the right medications. By swiftly detecting diseases and offering prompt remedies, our software aids in preserving the production of spinach and ensures farmers can effectively combat fungal diseases, ultimately benefiting the food, medicine, and skincare industries.

Capturing and Using Taxonomic Knowledge for Species Identification

The skills and knowledge needed to recognize and classify taxa are becoming increasingly scarce in the scientific community, resulting in a “taxonomic impediment”, where such knowledge is gradually disappearing (Engel et al. 2021). At the same time, it is clear that these skills are strongly needed in biodiversity monitoring for management and conservation, especially when carried out by citizen scientists. Formalizing the knowledge required to distinguish taxa from one another in the form of open digital identification keys is one way of making such knowledge more findable, accessible, interoperable and reusable (FAIR) for professional and amateur observers of biodiversity. Experience working with taxonomic experts to capture taxonomic knowledge in a tabular, multiple access format has proven that important nuances in taxonomic knowledge cannot be captured in a tabular format in a practical manner. Features like taxonomic hierarchy, rich metadata, formalized geographical scope, multilingualism, numerical ranges, logical premises, and links to external services are all much needed improvements of digital identification keys. However, these are complex to capture in a tabular format, defeating the benefits of its general simplicity. To address this, we have developed Clavis, an open format for identification keys written in JavaScript Object Notation (Koch et al. 2022). The goal is to create an ecosystem where different implementations and mappings of the Clavis format can serve the needs of various users and institutions, while maintaining knowledge exchange. Having a versatile format as a foundation, a crucial next step was to offer experts a tool to edit and test identification keys, and provide end-users with a tool to access this information. In this presentation, we show the Clavis format itself, as well as an online editing tool and a end-user interface, available on clavis.no. Providing a more advanced use case, we demonstrate how we implement Clavis in tandem with computer vision, providing users with a single tool to quickly and reliably identify taxa using different technologies.

Pychastic: Precise Brownian dynamics using Taylor-Itō integrators in Python

In the last decade, Python-powered physics simulations ecosystem has been growing steadily, allowing greater interoperability, and becoming an important tool in numerical exploration of physical phenomena, particularly in soft matter systems. Driven by the need for fast and precise numerical integration in colloidal dynamics, here we formulate the problem of Brownian Dynamics (BD) in a mathematically consistent formalism of the Itō calculus, and develop a Python package to assist numerical computations. We show that, thanks to the automatic differentiation packages, the classical truncated Taylor-Itō integrators can be implemented without the burden of computing the derivatives of the coefficient functions beforehand. Furthermore, we show how to circumvent the difficulties of BD simulations such as calculations of the divergence of the mobility tensor in the diffusion equation and discontinuous trajectories encountered when working with dynamics on S^2S2 and SO(3)SO(3). The resulting Python package, Pychastic, is capable of performing BD simulations including hydrodynamic interactions at speeds comparable to dedicated implementations in lower-level programming languages, but with a much simpler end-user interface.

Codebase release 0.2 for Pychastic

In the last decade, Python-powered physics simulations ecosystem has been growing steadily, allowing greater interoperability, and becoming an important tool in numerical exploration of physical phenomena, particularly in soft matter systems. Driven by the need for fast and precise numerical integration in colloidal dynamics, here we formulate the problem of Brownian Dynamics (BD) in a mathematically consistent formalism of the Itō calculus, and develop a Python package to assist numerical computations. We show that, thanks to the automatic differentiation packages, the classical truncated Taylor-Itō integrators can be implemented without the burden of computing the derivatives of the coefficient functions beforehand. Furthermore, we show how to circumvent the difficulties of BD simulations such as calculations of the divergence of the mobility tensor in the diffusion equation and discontinuous trajectories encountered when working with dynamics on S^2S2 and SO(3)SO(3). The resulting Python package, Pychastic, is capable of performing BD simulations including hydrodynamic interactions at speeds comparable to dedicated implementations in lower-level programming languages, but with a much simpler end-user interface.

Natural Gas Flaring Management System: A Novel Tool for Sustainable Gas Flaring Reduction in Nigeria

The use of hydrocarbon fuels increases with population growth and rising standards of living, and so does natural gas flaring. Natural gas flaring is both a waste of natural resources and a violation of Nigeria’s energy policy for sustainable development through natural gas conservation. However, it remains the most cost-efficient and effective associated natural gas (ANG) management option in developing countries such as Nigeria. The World Bank’s initiative to eliminate routine gas flaring by 2030 has increased the need to limit or eliminate routine gas flaring. Often, studies on natural gas utilisation techniques fail to consider the lack of practical tools that integrate economic, technical, and regulatory factors into a gas flaring management framework, and the intricacies of existing tools, which often come at the expense of simplicity to achieve real-time information output. This paper aims to establish a framework and ANG management tool to reduce regular gas flaring in Nigeria. This research established a management framework (using a flowchart decision tree) and models to provide a user-friendly ANG flaring tool (using a MATLAB graphical front end user interface with back-end ASPEN HYSYS thermodynamic models). This was combined with techno-economic models for liquefied natural gas, gas-to-methanol, and gas-to-wire ANG utilisation options. The tool was then tested with data obtained from Fields Y and X in the Niger Delta region of Nigeria. The results, considering both economic and technical factors, showed that the choice of liquefied natural gas for Field Y was best due to its proximity to the pipeline infrastructure and its cost-effectiveness, and the availability of a high-demand LNG market for that area. For Field X, gas-to-wire was best due to its proximity to the electrical grid and high electricity requirements for that area. Additional geographical profiles in West Africa and ANG utilisation alternatives were recommended for further investigation. This paper developed and validated a one-of-a-kind ANG flaring management tool that incorporates techno-economic analysis of selected ANG utilisation options to assist operators and investors in making more profitable investment decisions.

ReadBouncer: precise and scalable adaptive sampling for nanopore sequencing.

MotivationNanopore sequencers allow targeted sequencing of interesting nucleotide sequences by rejecting other sequences from individual pores. This feature facilitates the enrichment of low-abundant sequences by depleting overrepresented ones in-silico. Existing tools for adaptive sampling either apply signal alignment, which cannot handle human-sized reference sequences, or apply read mapping in sequence space relying on fast graphical processing units (GPU) base callers for real-time read rejection. Using nanopore long-read mapping tools is also not optimal when mapping shorter reads as usually analyzed in adaptive sampling applications.ResultsHere, we present a new approach for nanopore adaptive sampling that combines fast CPU and GPU base calling with read classification based on Interleaved Bloom Filters. ReadBouncer improves the potential enrichment of low abundance sequences by its high read classification sensitivity and specificity, outperforming existing tools in the field. It robustly removes even reads belonging to large reference sequences while running on commodity hardware without GPUs, making adaptive sampling accessible for in-field researchers. Readbouncer also provides a user-friendly interface and installer files for end-users without a bioinformatics background.Availability and implementationThe C++ source code is available at https://gitlab.com/dacs-hpi/readbouncer.Supplementary information Supplementary data are available at Bioinformatics online.

Discovery of Biomedical Databases Through Semantic Questioning.

Many clinical studies are greatly dependent on an efficient identification of relevant datasets. This selection can be performed in existing health data catalogues, by searching for available metadata. The search process can be optimised through questioning-answering interfaces, to help researchers explore the available data present. However, when searching the distinct catalogues the lack of metadata harmonisation imposes a few bottlenecks. This paper presents a methodology to allow semantic search over several biomedical database catalogues, by extracting the information using a shared domain knowledge. The resulting pipeline allows the converted data to be published as FAIR endpoints, and it provides an end-user interface that accepts natural language questions.

A Low-Cost IoT Based Buildings Management System (BMS) Using Arduino Mega 2560 And Raspberry Pi 4 For Smart Monitoring and Automation

This work presents an internet of things (IoT) based building management system (BMS) for monitoring, control, and energy management in buildings to provide an efficient way of energy utilization. Existing systems mainly provide monitoring of different parameters with limited controlling/automation functions. Existing solutions also do not provide automatic decision-making, advanced safety management, and resource tracking. However, the proposed system provides a comprehensive way of monitoring, controlling, and automatic decision making regarding different environmental and electrical parameters in buildings, i.e., temperature, humidity, dust, volt, etc., by using a low-cost wireless sensor network (WSN). The architecture of the proposed system consists of five layers and uses analog sensors which are connected to Arduino Mega 2560 microcontrollers for data collecting, NodeMCUs ESP8266 for wireless communication, Raspberry Pi4 microcomputers for decision making, and nod-RED dashboard which runs locally on a Raspberry Pi 4to provide a friendly end-user interface. The system also uses the Message Queuing Telemetry Transport (MQTT) communication protocol through Wi-Fi and completely relies on the local devices in the architecture and does not need cloud computing services. The proposed system provides two different kinds of automation, i.e., safety automation for the safety of different devices with advanced features, and energy automation. The proposed system is also able to provide humidity control inside a room and to track and count the available resources in any facility. The proposed system is low cost, scalable, and can be used in any building. Simulation results show that the proposed system is highly efficient.

A Temporal Case-Based Reasoning Platform Relying on a Fuzzy Vector Spaces Object-Oriented Model and a Method to Design Knowledge Bases and Decision Support Systems in Multiple Domains

Knowledge bases in complex domains must take into account many attributes describing numerous objects that are themselves components of complex objects. Temporal case-based reasoning (TCBR) requires comparing the structural evolution of component objects and their states (attribute values) at different levels of granularity. This paper provides some significant contributions to computer science. It extends a fuzzy vector space object-oriented model and method (FVSOOMM) to present a new platform and a method guideline capable of designing objects and attributes that represent timepoint knowledge objects. It shows how temporal case-based reasoning can use distances between temporal fuzzy vector functions to compare these knowledge objects’ evolution. It describes examples of interfaces that have been implemented on this new platform. These include an expert’s interface that describes a knowledge class diagram; a practitioner’s interface that instantiates domain objects and their attribute constraints; and an end-user interface to input attribute values of the real cases stored in a domain case database. This paper illustrates resultant knowledge bases in different domains, with examples of pulmonary embolism diagnosis in medicine and decision making in French municipal territorial recomposition. The paper concludes with the current limitations of the proposed model, its future perspectives and possible platform enhancements.

Implementation of machine learning techniques for depression in text messages: a survey

Depression is a disease or problem associated with high levels of stress seen in humans. It is uncomfortable in talking to parents, psychologists, and healthcare professionals in general. So a virtual platform is much more suitable for sharing your emotions, for example, a chatbot that provides the user with a comfort zone, acting as a friend or well-wisher. Extracting and identifying emotions from text messages to detect depressive mood is a challenging task because it involves removing natural language ambiguities. Over the past decade, researchers have proposed various state-ofthe- art methods for detecting depressive moods in text. This paper aims to analyze such methods and present a comparison based on detection accuracy. The virtual platform provides an end-user interface for communication. The system understands the meaning and context of a sentence using Natural Language Processing (NLP), word embedding, and machine learning techniques. NLP does the preprocessing and extracts the mental health-related keywords. Word embedding converts the extracted keywords into embedding vectors that can be understood by Machine learning algorithms, it can also analyze and extract users' feelings by examining and calculating levels of depression and classifying the user as depressed or not. This paper showed that the support vector machine is the preferred algorithm over other machine learning algorithms and provides higher accuracy.

Power Line Communication based SmartPlug Prototype for Power Consumption Monitoring in Smart Homes

In order to implement the concepts of smart grid and residential automation, this article presents a SmartPlug prototype for power consumption monitoring in smart homes. It is a device that, when coupled to the sockets, measures electrical voltage and current, and thereafter provides the power consumption to the consumer using the power line communication (PLC) technology. With the knowledge of the consumption, provided by the smart metering, of each outlet, the consumer is able to establish the concept of segmented energy consumption, which is an important allied of energy saving and power demanding. With commercial current and voltage sensors, the SmartPlug was built using the Arduino platform and the end user interface was created using the open source software ScadaBR. After performance evaluation in a laboratory, tests in real electrical plants were performed to validate the prototype, considering bit rates and signal spectrum in the PLC systems.

MSCAT: A Machine Learning Assisted Catalog of Metabolomics Software Tools.

The bottleneck for taking full advantage of metabolomics data is often the availability, awareness, and usability of analysis tools. Software tools specifically designed for metabolomics data are being developed at an increasing rate, with hundreds of available tools already in the literature. Many of these tools are open-source and freely available but are very diverse with respect to language, data formats, and stages in the metabolomics pipeline. To help mitigate the challenges of meeting the increasing demand for guidance in choosing analytical tools and coordinating the adoption of best practices for reproducibility, we have designed and built the MSCAT (Metabolomics Software CATalog) database of metabolomics software tools that can be sustainably and continuously updated. This database provides a survey of the landscape of available tools and can assist researchers in their selection of data analysis workflows for metabolomics studies according to their specific needs. We used machine learning (ML) methodology for the purpose of semi-automating the identification of metabolomics software tool names within abstracts. MSCAT searches the literature to find new software tools by implementing a Named Entity Recognition (NER) model based on a neural network model at the sentence level composed of a character-level convolutional neural network (CNN) combined with a bidirectional long-short-term memory (LSTM) layer and a conditional random fields (CRF) layer. The list of potential new tools (and their associated publication) is then forwarded to the database maintainer for the curation of the database entry corresponding to the tool. The end-user interface allows for filtering of tools by multiple characteristics as well as plotting of the aggregate tool data to monitor the metabolomics software landscape.

Sequential deep learning from NTSB reports for aviation safety prognosis

In this paper, we apply a set of data-mining and sequential deep learning techniques to accident investigation reports published by the National Transportation Safety Board (NTSB) in support of the prognosis of adverse events. Our focus is on learning with text data that describes the sequences of events. NTSB creates post hoc investigation reports which contain raw text narratives of their investigation and their corresponding concise event sequences. Classification models are developed for passenger air carriers, that take either an observed sequence of events or the corresponding raw text narrative as input and make predictions regarding whether an accident or an incident is the likely outcome, whether the aircraft would be damaged or not and whether any fatalities are likely or not. The classification models are developed using Word Embedding and the Long Short-term Memory (LSTM) neural network. The proposed methodology is implemented in two steps: (i) transform the NTSB data extracts into labeled dataset for building supervised machine learning models; and (ii) develop deep learning (DL) models for performing prognosis of adverse events like accidents, aircraft damage or fatalities. We also develop a prototype for an interactive query interface for end-users to test various scenarios including complete or partial event sequences or narratives and get predictions regarding the adverse events. The development of sequential deep learning models facilitates safety professionals in auditing, reviewing, and analyzing accident investigation reports, performing what-if scenario analyses to quantify the contributions of various hazardous events to the occurrence of aviation accidents/incidents.

Real time analysis of Intelligent placing system for vehicles using IOT with Deep learning

There is an enormous in number of vehicles in last few years. So, it becomes important to make effective use of technology to enable inconvenience free parking at public and/or private places. In traditional placing systems, drivers face difficulty in finding available placing slots. These systems ignore the fact of placing the vehicles on roads, time management in peak hours, wrong parking of a vehicle in a placing slot. Moreover, the traditional systems require more human intervention in a placing zone. To deal with above said issues, there is an urgent requirement of developing Real Time Intelligent placing system. In this proposed system a Real Time Intelligent placing system based on IOT and Deep learning techniques to answer the real time management of placing and uncertainties. The proposed solution utilizes smart sensors, cloud computing and cyber physical system. Development of graphical user interface for administrator and end-user is a major challenge as it requires ensuring smooth monitoring, control and security of placing system. Moreover, it needs to establish effortless coordination with an end-user. The proposed system is successful in smartly addressing the challenges such as indicating status of placing slot well in advance to end-user, use of reserved and unreserved placing slots, wrong placing, unauthorized placing, real time analysis of free and occupied slots, detecting multiple objects in a placing slot such as bike in car slot, fault detection in one or more components and traffic management during peak hours. The system reduces the human work, saves time, money and energy.

Perancangan Expert System Development Life Cycle pada Sistem Pakar Forward Chaining Sebagai Media Pembelajaran

The purpose of this research is to create and develop learning media in the form of applications used for plant identification in The Division of Pinophyta. The approach used is conventional approach with Expert System Development Life Cycle (ESDLC) method from Durkin, the inference mechanism used is Forward Chaining, while the software, used for Data Base Management System (DBMS) is MySQL, Unified Modeling Language (UML), Personal Home Page (PHP). Based on the results of the application assessment test in general obtained results that this expert system application deserves to be used as a learning media with criteria for Quality of end-user interface design (55.5%), Engagement (52.3%), Interactivity (64.3 %) and Tailorability (52.4%). Conclusion, application made by Expert System Development Life Cycle (ESDLC) method, with Forward Chaining inference mechanism worthy to be used as a learning medium of plant identification in Divisio Pinophyta.
 
 Keywords: Application, Plant Identification, Learning Media, Expert System, Forward Chaining

Unsupervised semantic and instance segmentation of forest point clouds

Terrestrial Laser Scanning (TLS) has been increasingly used in forestry applications including forest inventory and plant ecology. Tree biophysical properties such as leaf area distributions and wood volumes can be accurately estimated from TLS point clouds. In these applications, a prerequisite is to properly understand the information content of large scale point clouds (i.e., semantic labelling of point clouds), so that tree-scale attributes can be retrieved. Currently, this requirement is undergoing laborious and time consuming manual works. In this work, we jointly address the problems of semantic and instance segmentation of forest point clouds. Specifically, we propose an unsupervised pipeline based on a structure called superpoint graph, to simultaneously perform two tasks: single tree isolation and leaf-wood classification. The proposed method is free from restricted assumptions of forest types. Validation using simulated data resulted in a mean Intersection over Union (mIoU) of 0.81 for single tree isolation, and an overall accuracy of 87.7% for leaf-wood classification. The single tree isolation led to a relative root mean square error (RMSE%) of 2.9% and 19.8% for tree height and crown diameter estimations, respectively. Comparisons with existing methods on other benchmark datasets showed state-of-the-art results of our method on both single tree isolation and leaf-wood classification tasks. We provide the entire framework as an open-source tool with an end-user interface. This study closes the gap for using TLS point clouds to quantify tree-scale properties in large areas, where automatic interpretation of the information content of TLS point clouds remains a crucial challenge.

Impact Of ERP Implementation In Higher Education In India – A Review

ERP stands for enterprise resource planning. It is an integrated system through which an organization can manage the functioning of its various divisions whether logistics, production, finance, administration, finance and human resource. ERP has been widely used in various industry and businesses as it provides one application and single database to organize the entire enterprise. ERP system has been widely accepted and has shown positive response from the industry and business sectors. As a result, it has found its ingrowth in the ever-growing education sector ranging from schools to higher educational institutes like universities. The ERP system is being quickly adapted by the educational sector to stream line its functioning. But the ERP model for educational sector is completely different from that of business sector. The available ERP models include the sub-categories which are influenced by the needs of the industry sector. Majority of the educational institutes are re-engineering the business model of ERP for their educational system, which is not completely in-accordance to their requirements, hence rendering it unsuccessful. In the present work the authors have reviewed the literature in relation with the post-implementation consequences of ERP system in educational sectors worldwide. It has been observed that majority of organisations has re-engineered the already available ERP system to fulfil their purpose, which has rendered their manual to online transition an unsuccessful attempt. Also, the lack in training at the end-user interface of the software implementation is a major factor influencing the filed transition.

Machine Learning and IoT based Real Time Parking System: Challenges and Implementation

There is a tremendous increase in number of vehicles in last two decades. So, it becomes important to make effective use of technology to enable hassle free parking at public and/or private places. In traditional parking systems, drivers face difficulty in finding available parking slots. These systems ignore the fact of parking the vehicles on roads, time management in peak hours, wrong parking of a vehicle in a parking slot. Moreover, the traditional systems require more human intervention in a parking zone. To deal with above said issues, there is an urgent requirement of developing Smart Parking Systems. In this manuscript, the authors propose a Smart Parking System based on IoT and Machine learning techniques to answer the real time management of parking and uncertainties. The proposed solution utilizes smart sensors, cloud computing and cyber physical system. Development of graphical user interface for administrator and end-user is a major challenge as it requires to ensure smooth monitoring, control and security of parking system. Moreover, it needs to establish effortless coordination with an end-user. The proposed system is successful in smartly addressing the challenges such as indicating status of parking slot well in advance to end-user, use of reserved and unreserved parking slots, wrong parking, unauthorized parking, real time analysis of free and occupied slots, detecting multiple objects in a parking slot such as bike in car slot, fault detection in one or more components and traffic management during peak hours. The system minimizes the human intervention and saves time, money and energy.