Interpretable AI models have emerged as crucial tools for promoting transparent decision-making in complex data science scenarios. As artificial intelligence continues to permeate various industries, the need for models that can provide clear explanations for their decisions has become increasingly apparent. This paper outlines the significance of interpretability in AI models and highlights the challenges posed by opaque systems in handling intricate data science scenarios. We discuss various approaches and techniques aimed at enhancing interpretability, including feature importance techniques, surrogate models, local explanations, and simplified models. Moreover, we emphasize the importance of transparent decision-making in critical domains such as healthcare, finance, and criminal justice, where the consequences of AI-driven decisions can be profound. Through case studies and literature review, we elucidate the benefits and limitations of interpretable AI models and propose future research directions in this field. Our findings underscore the importance of interpretable AI models in fostering trust, accountability, and regulatory compliance, while also acknowledging the trade-offs between interpretability and performance. Overall, this paper provides insights into the role of interpretable AI models in enabling transparent decision-making and lays the groundwork for further advancements in this critical area of research.
The convergence of Artificial Intelligence (AI) and Big Data research has catalyzed unprecedented advancements across diverse sectors, revolutionizing the way data is analyzed, interpreted, and utilized. However, this rapid progress brings to the forefront a myriad of ethical considerations, particularly concerning privacy rights and individual autonomy. This paper delves into the intricate intersection of AI-enabled Big Data research and ethical considerations, aiming to strike a delicate balance between fostering innovation and safeguarding privacy. Ethical frameworks provide the foundational principles guiding researchers and practitioners in maleficence, and justice underscore the importance of prioritizing societal welfare, minimizing potential harms, and ensuring equitable access and distribution of benefits. These frameworks serve as ethical compasses, guiding researchers towards responsible and ethical conduct throughout the research process. Privacy concerns loom large in AI-enabled Big Data research, fueled by the unprecedented scale, scope, and granularity of data being collected and analyzed. The identification, aggregation, and inference of sensitive information from vast datasets raise significant privacy risks, challenging traditional notions of privacy protection. Moreover, the opacity of AI algorithms and the lack of transparency in decision-making processes exacerbate privacy concerns, undermining individuals' ability to understand and control the use of their personal data. Regulatory approaches play a crucial role in addressing ethical concerns in AI-enabled Big Data research, providing a framework for legal compliance and accountability. Regulations such as the General Data Protection Regulation (GDPR) and the California Consumer Privacy Act (CCPA) impose obligations on organizations regarding data collection, processing, and consent, aiming to empower individuals with greater control over their personal data. However, regulatory frameworks must evolve in tandem with technological advancements and emerging ethical challenges, ensuring effective protection of privacy rights in the digital age. Emerging technologies offer promising solutions to mitigate ethical concerns in AI-enabled Big Data research while enabling innovation to flourish. Techniques such as differential privacy, federated learning, and explainable AI enhance privacy preservation, transparency, and interpretability of AI systems, fostering trust and accountability. By leveraging these technologies, researchers can uphold ethical principles while harnessing the transformative potential of AI and Big Data for societal benefit.
In the rapidly evolving landscape of data science, predictive modeling stands as a cornerstone for deriving actionable insights from vast amounts of data. Central to the success of predictive modeling is the process of feature engineering, which involves selecting, transforming, and creating features to improve model performance. With the advent of artificial intelligence (AI) and machine learning (ML), automated feature engineering has emerged as a promising approach to streamline and enhance this critical process. This paper explores the role of AI-driven automated feature engineering techniques in augmenting the performance of predictive models in data science.The paper begins with an overview of predictive modeling in data science, highlighting the significance of feature engineering in model development. [1] Traditional approaches to feature engineering often rely on manual experimentation and domain expertise, which can be time-consuming and prone to human bias. In contrast, AI-driven automated feature engineering leverages ML algorithms and techniques to automate and optimize the feature engineering process, reducing the need for manual intervention and accelerating model development. Various AI-driven automated feature engineering techniques are examined, including machine learning-based feature selection algorithms, automated feature transformation methods, generative adversarial networks (GANs) for feature creation, and deep learning-based feature extraction techniques. These methods offer advantages such as improved model performance, time and resource efficiency, and reduced human bias. The paper also discusses challenges and limitations associated with AI-driven automated feature engineering, such as data quality requirements, interpretability of automated features, and the risk of overfitting. Additionally, case studies and applications demonstrate the practical utility of automated feature engineering across diverse domains, including finance, healthcare, marketing, and more. The paper explores future directions in automated feature engineering, including emerging trends, integration of domain knowledge, and ethical considerations. By providing valuable insights into the methodologies, tools, benefits, challenges, and future prospects of AI-driven automated feature engineering, this paper aims to guide practitioners and researchers in harnessing advanced techniques to enhance predictive modeling in data science.
With the uprising cases of corona virus (COVID-19), the whole world has come to stand still The deadly virus has infected more than 672,000 cases and caused 31,000 deaths and the number is still rising The intensity of pressure on government is too high, so it is critical to understand the reason of rising number and ways it can be controlled The study shows the analysis of the spread of COVID-19 on the basis of age, travel history and symptoms-based clusters In this paper, data of Ministry of Health (MoH) of the COVID -19 affected country China, till 29h March 2020, is analysed based on basic demographic data (age, sex and symptoms) Clustering is used to categorise the symptoms based on age group ranging from 0 to 87 years © 2020 SERSC
This paper presents the assessment model of competence certification for construction workers.Based on the Law of the Republic of Indonesia number: 2 years 2017 on Construction Service that the construction worker in the field of construction Services must have a certificate of competence work.In the implementation of the certification assessment of the competence for construction workers within the Construction Service Development Board (CSDB) required standard norms or guidelines used by the competency assessor of work in the test or assess the applicant certification of competence.The research model used in this study is a systematic review consisting of 4 stages, namely Problem formulation, Search literature, literature selection, analysis and interpretation.Assessment model of competence certification consists for construction workers, test material, assessment tool, usage manual and assessment guide.With this assessment model assessors will be more helpful so that the process of competence certification for construction workers become more optimal and produce a quality work.
Spray drying technology is an effective postharvest process for making fruit and vegetable powders with long shelf life.In general, spray dryer machine needs electricity energy for more than 5KW which can't be afforded by farmer community.In this study, low electricity energy spray drying machine was designed for farmer community.The machine use food grade stainless steel with the combination of electrical energy and liquid petroleum gas (LPG) energy.Electrical energy is used for electronics control system and air compressor, whereas LPG energy use for heat exchanger for the drying chamber.The electronics process controller is implemented to convert fresh fruit or vegetable into a fruit or vegetable powder form of a desired quality at a minimum cost.A simulation of a PID controller with Ziegler-Nichols tuning in the form of an Excel spreadsheet give the parameters used for the machine.Experiment was done to produce tomato powder from fresh tomato, whereas air inlet temperature of 150℃, the pressure of 3 bar and hot air outlet set to 100℃ with the addition of Maltodextrin 40mL/Liter has increase the powder quality, however adding more Maltodextrin make the tomato powder has less natural flavor.Increasing hot air temperature make the powder quality increase, in contrary the energy consumption of making the powder increased.
A large number of Wireless Sensor Network (WSN) devices are used to monitor the data from the environment and many applications such as disaster management, monitoring, military, healthcare and security etc. are used to monitor the environment.In these networks sensor nodes transmission rate is high and it leads to limit the battery power.This causes network redundancy and network life time will not be prolonged.In this paper, this can be addressed by using with the help of agents.Here, agents are used to collect data from the cluster head instead of forwarding directly to the mobile sink.When data collection is done, agents will transmit data to the mobile sink.This may reduce energy consumption and prolongs the network life in WSNs.Along with agents, existing techniques are also presented in this paper with comparative study.
In the business context, the internet brings with it a transformational impact that creates a new paradigm in business, digital marketing.With the use of the internet mobile, facilitate every transaction to be done by a person like ordering food, clothing purchases, purchasing household needs, and others.Everything is facilitated by the e-commerce.For tourist, traveling is one of necessity, still however there are other requirement that is needed is social.Jakarta Hidden Tour showcases another side of Jakarta city that invites to see, feel, and experience some part of the city to discover the intercultural meeting point of view.This paper presents E-Commerce website transformation especially to be applied for private travel agent.It explores the transformation of the website into e-commerce at Jakarta Hidden Tour, studies the system from e-commerce Jakarta Hidden Tour and formulates future development for ease of transaction.The results show that the Jakarta Hidden Tour starts from using platforms WordPress to BlogSpot because it is free and easy to use.The e-commerce system is then changed to online data.Future development plans can adapt to an increasingly advanced era of application creation, online booking, online transaction or payment as well as development of a tour destination.
Volcanic debris flow disaster is highly triggered by rainfall.However, limited access to the area of active volcano slope and damage of the observation station restricts the direct measurement by rain gauges.High-resolution X-band weather radars have been extensively used in hydrological researches and flood mitigation programs.In this study, the potential utilization of X-band multi-parameter compact (X-MP) radar for volcanic disaster mitigation is in real-time is presented.The study area is the rivers on Mount Merapi, which is historically the most active volcano in Indonesia.In the first part, the use of X-MP radar in real-time scheme is described.This part demonstrates the radar-rainfall estimation and the first attempt to predict the rain echo motion in short-term by using extrapolation model.In the second part, the advantage of radar for showing the spatially predominant rainfall for vulnerability assessment is demonstrated.The susceptibility level of three river basins, Pabelan River, Boyong River, and Gendol River, is generated using radarrainfall spatial distribution intensive observation period between October 2015 and February 2016.The real-time analysis has shown the advantage of radar to observe short-localized rainfall event.The results of radar extrapolation model suggest the consideration of uncertainties in the prediction system.The susceptibility score calculated from frequency of rainfall threshold exceedance and slope calculated by parametric modeling technique can be used to determine the susceptible area.The analysis finds that generally Boyong River is the most prone area for lahar flow, particularly at the region within 2 km to 3 km from the summit.The proposed X-MP radar utilization would be useful for mitigation of multimodal sediment disaster caused by volcanic eruption.
HOV lanes are supposed to control the traffic flow and keep it uncongested; nonetheless, HOV lanes require the continuation of the traffic jams and congestions; otherwise they will have no effect.The more traffic jams and congestions take place, the more successful the HOV lanes will be.The absurdity of HOV is that a solution for traffic congestions has a critical need for constant congestions to be successful.In the coming years when autonomous vehicle will be more common, the effectiveness of HOV lanes will be even slighter, but even before the emerging of the autonomous vehicles, the concept of HOV lanes failed in many venues because its reasoning is irrational
In Mobile Ad hoc Networks, routing is a challenging issue.Basically, the working routing protocols for these networks are classified in to three categories like the reactive protocols, proactive protocols and the hybrid t protocols which were the combination of above two types of protocols.Among them proactive routing protocols category is selected for the present study.In proactive routing protocols, STAR with LORA approach is chosen because it has less control overhead when compared to ORA approach and other proactive routing protocols.It works on few defacto parameter values in a dynamic MANET environment.These static values for parameters are not suitable in a dynamic environment.With reference to the IETF draft, it is a time series problem.In this paper, an effort has been made to incorporate the soft computing technique, fuzzy logic based STAR to enhance the performance of a MANET to support real time communication.The proposed Fuzzy logic approach based STAR performance is evaluated using simulation through QualNet simulator.From the results, it is observed that the Fuzzy logic based STAR provides superior performance than the defactoSTAR protocol.The comparative performance was measured using the performance metrics End-to-end delay, Jitter and throughput.A number of simulation scenarios were executed for small, medium and large size networks.From the simulation results, we conclude that Fuzzy Logic based STAR outperforms for small, medium size networks.
Through numerous vehicle technologies advances, vehicular ad hoc network (VANET) applications have emerged as a new paradigm for the automotive industries.VANET is a special kind of mobile ad hoc network (MANET) which is a revolutionary technology that allows vehicles to be interconnected with each other using VANET communications protocols such as vehicle to vehicle (V2V), vehicle to infrastructure (V2I), vehicle to pedestrian (V2P), and vehicle to network (V2N).From this communication infrastructure, VANET can provide various services to the connected vehicle and their users.However, each communication protocol cannot interoperate with one another due to their characteristics.Although much research has been proposed to resolve the issues, some important issues have not yet been addressed.To resolve service silo, we provide an ontology for integrating existing vehicle services.To this end, we define the characteristics of each protocol and represent each resource in the ontology.Finally, to show feasibility, we implement the ontology using the protégé tool.Through the test results of this implementation, we can be sure that the vehicle ontology can eventually contribute to integrating various vehicle services in vehicle to everything (V2X) communication.
Frequently, the neutron transmission through the shields used for protection against radiation is an unavoidable phenomenon, so we have interested in this work to study the neutron transmission through shields.We have considered an infinite homogenous slab witch characterized by his scattering probability noted Ps, with a different thickness and an infinite plane source of neutrons which arrived on the left side of the slab and on the right side detector with fixed window is placed to detect transmitted neutrons and evaluate the neutron transmission probabilities.We used the simulation Monte Carlo method for sampling the neutron history in the slab and in order to accelerate the calculation convergence we have developed a new multi-parameters spatial biasing technique with 4 parameters.For each thickness of the slab and for several values of Ps we have determined the detector response and calculated the neutron transmission probability.We compared our result by results obtained with the spatial biasing technique with 2 parameters and 1 parameter.Then we have determined the FOM (Figure of the Merit) for each method.We can also notice that our method presents bests results by obtaining the greatest FOM for a large thickness of the slab having high scattering probability Ps.
Proportional plus Resonant controller is presented as the injected grid current regulator applied to a single-phase grid connected inverter.This paper establishes a systematic approach to design the gains of the PR controller by determining its minimum and maximum threshold points, to achieve improved transient and steady-state performance.Extensive simulations are carried out by varying system parameters and their effects on magnitude and bandwidth of the amplitude-frequency response are explained.
Text classification for data preprocessing methods regularly uses bagof-words (BoWs).In a large dataset, BoWs always include many vectors with very large sizes and high dimensions.The authors introduced a new data preprocessing method for feature reduction of short text classification, namely NDTMD.It reduces features of the dataset using BoWs and word embedding (WE), and can solve the weaknesses of BoWs.The experiment consisted of four steps: 1) 5 datasets were selected from the data science community website, Kaggle; 2) the new methods were compared with 5 commonly used data preprocessing methods and 4 of these 5 methods used the state of the art as their baseline, while the other one used BoWs.One of the new data preprocessing methods used features reduction of BoWs to produce a new document termed matrix data (NDTMD); 3) the authors generated many classification models by 3 classifiers: support vector machine, logistic regression, and convolutional neural network for text classification; and 4) the above classifiers were applied to each preprocessing dataset and evaluated using feature reduction rate (FRR), accuracy, kappa, and running time performance.The results showed that classification models had the highest performance when using NDTMD.In particular, classifier algorithms had the highest accuracy and kappa but the lowest running time.The new data preprocessing methods can be used to preprocess short text classification and also can be applied with real social media data.
In this current industrial generation, accustoming to Industry 4.0 or 4th generation automation has been widely recognized.Converting traditional to ergonomically sophisticated machines to function automatically without any intervention of humans have become available in the market.Due to its assurance of efficiency, quality, reliability, and superiority over manual systems, incorporating Industrial Automation has improved productivity within a short period of time.Consequently, upgrading automated machines by integrating a programmable logic controller (PLC) has been a promising solution.Furthermore, this PLC is equipped with a human-machine interface (HMI) which creates a new connectivity method called the Industrial-Internet of Things (I-IoT).This paper focuses on consolidating a completely automated hydraulic machine, the features of IoT and implementing Machine-to-Machine (M2M) Technology.This achieves the main objective of industrial energy saving, occupational safety, labor extensiveness and lowering time consumption by establishing remote control and monitor, remote troubleshoot, feasible technological human interaction and predictive maintenance.
In present scenario, the power system is becoming extensive and more multifaceted.It is important to predict the line flows and bus voltages for dissimilar operating circumstances and network topologies of a power system.Failure of any equipment's during its operation harms the reliability of the system, hence leading to outages.Cascading outages can have a catastrophic influence on power system security.Ensuring the safe and reliable operation of a power system requires assessing both the static and the dynamic operations of the power system.Transmission switching is conveyed as an optimization problem to regulate the most influential lines as candidates for the interruption.The off line analysis to predict the effect of individual incident is a tedious task as a power system contains a large number of components.A detailed security assessment is essential to deal with the possible failures in the system, its consequences and its remedial actions.The proposed method has been tested on IEEE39 bus system with various loading and outage conditions using the Mi -Power.
With a sponsored content plan on the Internet market, a content provider (CP) negotiates with the Internet service providers (ISPs) on behalf of the end-users to remove the network subscription fees. In this work, we have studied the impact of data sponsoring plans on the decision-making strategies of the ISPs and the CPs in the telecommunications market. We develop game-theoretic models to study the interaction between providers (CPs and ISPs), where the CPs sponsor content. We formulate the interactions between the ISPs and between the CPs as a noncooperative game. We have shown the existence and uniqueness of the Nash equilibrium. We used the best response dynamic algorithm for learning the Nash equilibrium. Finally, extensive simulations show the convergence of a proposed schema to the Nash equilibrium and show the effect of the sponsoring content on providers’ policies.
This study analyzes the difference of Operation & Maintenance (O&M) costs between government's budgeting method and an alternative approach using Analytical Hierarchy Process (AHP).The AHP was used to determine the formula of budget distribution by considering irrigation infrastructure, location of irrigation area.The formulation was calculated based on the observation to samples in West Java, Central Java and East Java, three provinces that have significant agricultural contributions in Indonesia.Irrigation areas characteristics assessed were areas, channels, weirs and hydraulic structures.The result shows that the equation had an exceptional different linier relationship compared with the quadratic relation equation.They were far more reasonable having a linier proportionality between the budget and the amount of infrastructure.The conclusion also integrates with the linier relationship.Thus, while the precise equation are not intended to be universal applicable, the most important finding was how O & M budget was correlated positively to the amount of infrastructure, not only depending on the area of irrigation.
