Search

Accurate prediction of Dissolved Oxygen (DO) is an integral part of water resource management. This study proposes a novel approach combining Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) with AdaBoost and deep learning for multi-step forecasting of DO. CEEMDAN generates Intrinsic Mode Functions (IMFs) with different frequencies, capturing non-linear and non-stationary characteristics of the data. The high-frequency and medium-frequency IMFs, characterized by complex patterns and frequent changes over time, are predicted using Adaboost with Bidirectional Long Short-Term Memory (BiLSTM) as the base estimator. The low-frequency IMFs, characterized by relatively simple patterns, are predicted using standalone Long Short-Term Memory (LSTM). The proposed CEEMDAN-AdaBoost-BiLSTM-LSTM model is tested on data from ten stations of river Ganga. We compare the results with six models without decomposition and four models utilizing decomposition. Experimental results show that using a tailored prediction technique based on each IMF's distinctive features leads to more accurate forecasts. CEEMDAN-AdaBoost-BiLSTM-LSTM outperforms CEEMDAN-BiLSTM with an average improvement of 25.458% for RMSE and 37.390% for MAE. Compared with CEEMDAN-AdaBoost-BiLSTM, an average improvement of 20.779% for RMSE and 28.921% for MAE is observed. Diebold-Mariano test and t-test suggest a statistically significant difference in performance between the proposed and compared models. [ABSTRACT FROM AUTHOR]

A novel infectious coronavirus disease (COVID-19) identified in late 2019 has now been labelled as a global pandemic by World Health Organization (WHO). The COVID-19 outbreak has shown some positive impacts on the natural environment. In present work, India is taken as a case study to evaluate the effect of lockdown on air quality of three Indian cities. The variation in concentration of key air pollutants including P M 10 , P M 2.5 , N O 2 , S O 2 and O 3 during two phases, pre-lockdown and post-lockdown phases, was analysed. The concentration of P M 10 , P M 2.5 , N O 2 and S O 2 reduced by 55%, 49%, 60% and 19%, and 44%, 37%, 78% and 39% for Delhi and Mumbai, respectively, during post-lockdown phase. Overall, the findings in present study may provide confidence to the stakeholders involved in air quality policy development that a significant improvement in air quality can be achieved in future if better pollution control plans are strictly executed. [ABSTRACT FROM AUTHOR]

Reservoir characterization is one of the most challenging tasks that help in modeling different lithological properties like porosity, permeability and fluid saturation using seismic readings like velocity profile, impedance, etc. Such a model is required for field development, placing new wells and prediction management. Seismic attributes are being progressively utilized for the tasks of model building, exploration and properties estimation from the data. However, these tasks become very complex due to the nonlinear and heterogeneous nature of subsurface properties. In this context, present work proposes a recurrent neural network-based learning framework to classify porosity using seismic attributes as predictor variables. The approach begins by calculating different seismic attributes from the data. From the initially calculated attribute set, features that are to be used for classification are selected by using two different strategies. Firstly, the seismic attributes having good correlation strength with reservoir porosity are extracted. Subsequently, generative topographic map is utilized to select the significant features. The final reduced features set obtained by the integrated result of above two strategies is then fed as an input to the empirical mode decomposition (EMD) algorithm. The denoised features resulting from the EMD algorithm are used to train the classification models. Further, a comparison is carried out between the proposed classification framework (E M D + R N N) and other supervised classifiers to show the performance of the proposed framework. [ABSTRACT FROM AUTHOR]

Enabling deep neural networks for tight resource constraint environments like mobile phones and cameras is the current need. The existing availability in the form of optimized architectures like Squeeze Net, MobileNet etc., are devised to serve the purpose by utilizing the parameter friendly operations and architectures, such as point-wise convolution, bottleneck layer etc. This work focuses on optimizing the number of floating point operations involved in inference through an already compressed deep learning architecture. The optimization is performed by utilizing the advantage of residual connections in a macroscopic way. This paper proposes novel connection on top of the deep learning architecture whose idea is to locate the blocks of a pretrained network which have relatively lesser knowledge quotient and then bypassing those blocks by an intelligent skip connection, named here as Shunt connection. The proposed method helps in replacing the high computational blocks by computation friendly shunt connection. In a given architecture, up to two vulnerable locations are selected where 6 contiguous blocks are selected and skipped at the first location and 2 contiguous blocks are selected and skipped at the second location, leveraging 2 shunt connections. The proposed connection is used over state-of-the-art MobileNet-V2 architecture and manifests two cases, which lead from 33.5% reduction in flops (one connection) up to 43.6% reduction in flops (two connections) with minimal impact on accuracy. [ABSTRACT FROM AUTHOR]

Owing to accurate future air quality estimates, need for detecting the anomalously high increase in concentration of pollutants cannot be adjourned. Plentiful approaches were proposed in the past to substantially determine the abnormal conditions, but most of the statistical approaches were computationally expensive and ignored the false alarm ratios. Thus, a hybrid of proximity- and clustering-based anomaly detection approaches to identify anomalies in the air quality data is suggested in this work. The Gaussian distribution property of the real-world data set is utilized further to segregate out anomalies. The results depicted twofold advantages of our approach, by efficient extraction of anomalies and with increased accuracy by reducing the number of false alarms. Specifically, the presence of NO2 concentration in air is investigated in this work, considering its constant increase over decades as well as its inevitable health risks. Furthermore, spatiotemporal segments with anomalously high NO2 concentrations for 14 residential, industrial, and commercial areas of five cities in India are extracted. To validate the results, a comparative analysis with existing approaches of anomaly detection and with two benchmark data sets is performed. Results showed that our method outperformed the existing methods of anomaly detection, when evaluated over metrics such as sensitivity, miss rate, and false alarms. Further, a detailed analysis of extracted anomalies and a detailed discussion about the factors responsible for such anomalies are presented in this work. This study is helpful in educating government and people about spatiotemporal, geographical, and economic conditions responsible for anomalously high NO2 concentrations in air. Implications: Using our methodology, days with extremely high concentration of any pollutant in air, at any particular location, can be extracted. The reasons for such extremely high pollutant concentration on particular days of a year can be studied and preventive measures can be taken by the government. Thus, by identification of causes of anomalies, future similar events can be avoided. This would also help in people's decision making in case such events occur in the future. [ABSTRACT FROM AUTHOR]

Seismic attributes are being increasingly utilized in both reservoir characterization and exploration. In the Reservoir characterization, one might be interested in modeling different reservoir properties like water saturation, permeability, and porosity. The direct estimation of these reservoir properties from the seismic data is a challenging and time-consuming task. Therefore, it is desirable to build a model that uses seismic attributes to measure the different lithological properties of need. In recent decades, various advancements have been done to estimate porosity from seismic attributes and well-logs effectively. However, the modeling of nonlinear porosity patterns is still underdeveloped for robust solutions as the existing methods implement linear or functional network based reduction techniques to estimate petrophysical properties. Therefore, in this study, we propose a nonlinear hybrid approach that integrates Generative Topographic Map with Machine Learning prediction models (ML) to estimate the property porosity. The approach initiates by extracting relevant attributes based on the two different feature selection strategies. Subsequently, the Radial Basis Function (RBF) and Random Forest regressors are used to build the training model for the target well property porosity using extracted attributes. The different combinations of wells are formed to show the variations in model training and testing error. To demonstrate the effectiveness of proposed PP-NFR approach, the prediction results of the proposed hybrid approach are compared with the linear prediction technique. Furthermore, the performance of PP-NFR approach is also evaluated by comparing the prediction results with the state-of-the-art prediction models (ANN and SVM). The prediction results revealed that the proposed non-linear hybrid approach outperforms other state-of-the -art prediction models by providing an average prediction error of 0.124 as compared to 0.15 and 0.159 generated by SVM and ANN models respectively. • In the present work, we propose a nonlinear hybrid approach to predict porosity from seismic attributes and target well-log. • The approach combines nonlinear reduction technique with machine learning models to support improved prediction accuracy. • The performance evaluation of proposed approach is done by comparing prediction results with state-of-the-art prediction models. • The proposed approach generates an average prediction error of 0.124 as compared to 0.15 and 0.159 generated by SVM and ANN models respectively. [ABSTRACT FROM AUTHOR]

Highlights • Enabling Radial Basis Function for surface fitting using Big Data and limited memory. • Solution for missing value prediction on Big Data. • Performance for missing value prediction is found better than Kernel Ridge Regression. • Establishes bias variance tradeoff with the help of hyperparameters. Abstract The problem of missing values in the process of data acquisition is becoming critical due to both hardware failure and human error. Radial Basis Function based interpolation and missing value prediction in the dependent variable through surface fitting has been the viable solution from a long time. However, the solution works well on the idea of building one equation of 'N' weight variables corresponding to each sampled data point from a set of 'N' samples. Still, because of the inherent computations, it demands big primary memory when 'N' becomes large. Hence, on a memory-restricted setup, sometimes it suffers from the memory overflow problem. In this paper, we propose a novel data decomposition based RBF enabled surface fitting approach by building and aggregating multiple small models in an overlapping manner, which works well even with smaller primary memory with minimal impact on loss of generality. We also consider two hyperparameters as window size and overlapping index in order to tune bias-variance tradeoff. The proposed approach is applied to ten real-world datasets having multiple dimensions and results are found competitive while comparing with single trained model and Kernel Ridge Regression. Therefore, we believe this approach will rejuvenate the RBF based surface fitting method in attaining better performance in the Big data world too. [ABSTRACT FROM AUTHOR]

Sustainable development of the national food system must ensure the introduction of adequate food security interventions and policies. However, several high-end technological developments remain unexplored, which can be used to gain explicit information regarding agricultural problems. In this direction, the presented work proposes AgrIntel, a framework consisting of multiple AI-based pipelines to process nationwide farmers' helpline data and obtain spatiotemporal insights regarding food-production problems on an extensive scale. AgrIntel overcomes several limitations of the existing methods used for similar objectives, including limited scalability, low frequency, and high cost. The call-logs dataset used in the study is obtained from the nationwide network of farmers' helpline centers, managed by the Ministry of Agriculture & Farmers' Welfare, Government of India. The article demonstrates the Spatio-temporal profile of one of India's highest food grain-affecting diseases, i.e., "blast in rice crop", to demonstrate the utility of the AgrIntel pipelines. First, the proposed framework extracts and clusters the precise geographical locations of farmers calling for help corresponding to the target agricultural problem. Next, the temporal modeling of the problem helps extract the critical dates corresponding to the crop disease/pest spread. Furthermore, by incorporating the historical agroclimatological data, the article introduces a new medium to extract the favorable weather conditions corresponding to the targeted disease/pest outbreak. In addition, the study explores the potential of Deep Learning models (based on Artificial Neural Network, Convolutional Neural Network, Gated Recurrent Unit and Long short-term memory unit) to efficiently predict the futuristic demand for assistance regarding target problems (RMSE of ≈ 1.5 and MAE of ≈ 0.9 query calls). The obtained results expose unrevealed insights regarding food production problems, significantly boosting the food security policy-designing procedure. • Extracting nationwide disease/pest hot spots using helpline data. • Obtaining weather parameters responsible for the outbreak of disease/pests. • Locating regions with increasing/disease agricultural problems in India. • Deep Learning-based models for forecasting farmers' demand for assistance. • Four AI-based pipelines for nationwide profiling of plant protection-related problems. [ABSTRACT FROM AUTHOR]

Objective: Powered Two Wheeler (PTW) vehicles are one of the preferred modes of transport used in India. Also, PTWs accidents are comparatively more frequent than other type of accidents on road. The influencing factors of PTW accidents are also differ from factors that affect other accident types. The objective of this study is to analyze newly available PTWs road accident data from Uttarakhand state in India and revealing the factors that affect the severity of these accidents in various districts of Uttarakhand.. Methodology: To analyze the factors that affect the severity of road accidents in Uttarakhand, initially we have compared three popular classification algorithms i.e. decision tree (CART), Naïve Bayes and Support vector machine on PTW accident data set. The decision tree algorithm's (CART) classification accuracy was found better than other two techniques. Hence we have preferred CART algorithm to extract the factors that affect the severity of PTWVs accidents in whole Uttarakhand state and its 13 districts separately. Experimental Results: The analysis of PTWVs accident data using CART for 13 districts of Uttarakhand and the whole state reveals that every districts have different factors associated with PTW accidents severity. There are some districts in Uttarakhand state which have similar PTW accident patterns, whereas few districts are found to have different PTW accident patterns. These results are very useful to understand the pattern of PTW accidents in Uttarakhand state. These results can certainly be helpful to overcome the PTWs accident rate in Uttarakhand state. [ABSTRACT FROM AUTHOR]

Class imbalance learning is the study of problems in which some classes appear more frequently than the others. Most existing works that study this problem assume data set to be dense and do not exploit the rich structure of the data. One such structure is the sparsity. In the present work, we focus on solving the class imbalance problem under the sparsity assumption. More specifically, a well-known Gmean metric for class imbalance learning problem in binary classification setting has been maximized, which results in a non-convex loss function. Convex relaxation techniques are used to convert the non-convex problem to the convex problem. The problem formulation in the present work uses L 1 regularized proximal learning framework and is solved via accelerated-stochastic-proximal gradient descent algorithm. Our aim in the paper is to show: (i) the application of proximal algorithms to solve real world problems (class imbalance); (ii) how it scales to Big data; and (iii) how it outperforms some recently proposed algorithms in terms of Gmean, F-measure and Mistake rate on several benchmark data sets. [ABSTRACT FROM AUTHOR]

The government schemes (also known as programs and plans) or social welfare policies can be defined as the set of assistance and aids provided by the country's governance body. These schemes focus on the improved well-being of needful citizens. Some researchers have shown that introducing such policies and schemes has had an electoral impact in democratic countries. These earlier studies relied upon the post-poll and public survey data to reach conclusions. However, this data source has limitations and has to be collected manually, which makes it time-consuming and costly. The readily available internet inculcates the sharing of opinions freely on social media, facilitating government–citizen interactions. These interactions may show fluctuations in frequency and intensity on social media with the success and failure of some government schemes. Thus, this research proposes utilizing the Twitter data related to the government welfare schemes during the election duration to uncover the spatial and temporal relationships between the tweets' information diffusion pattern and political elections. To start with, we perform tweet classification to identify the target communities or groups and multiple user-engagements by employing deep learning-based pre-trained language representation (LR) models. The scarcity of labeled data limits the application of the supervised classification models on real-time data. Thus, we propose Mod-EDA, a text augmentation method to upscale the labeled data for reduced overfitting. Going further, we propose two modules, where the classified tweets are studied to investigate the scheme tweets' information diffusion pattern in correspondence to the election duration in terms of the voting phase and the electing parties , respectively. The proposed framework is evaluated for a case study of the 2019 Indian general elections. This study depicts that the voting phases and election duration trigger high government schemes related tweet generation. However, it is not affected by the location of the voting phase. The generation of complaints and negative tweets in one voting phase is covered with the positive news in subsequent voting phases. It is also seen that there is a strong influence of the ruling party on the scheme-related Twitter data generation. • Most of the government scheme tweets are generally promotional and informational. • Elections induce a high government scheme related tweet concentration. • The government scheme tweet concentration reduces just after the voting phase. • The type of information shared keeps changing as per the requirement. • The ruling party has a high regulation on the government scheme related tweets. [ABSTRACT FROM AUTHOR]

Seismic facies analysis provides an efficient way to identify the structure and geology of reservoir units. In the past two decades, seismic attributes have been widely used to highlight the geological features of interest. The combined benefits of multiple seismic attributes have been explored to estimate seismic events accurately. However, with the increase in the number of seismic attributes and size of seismic data, the traditional methods of seismic facies interpretation become very difficult and time-consuming. To address this issue, several computer-assisted facies classification algorithms including k-Means, self-organizing map and generative topographic map were introduced to capture the spatial distribution of seismic facies. The automated facies classification methods accurately quantify the characteristics of geological facies. However, these methods are computationally very time-expensive. To resolve this problem, we provide Spark Map-Reduce based framework for unsupervised classification of seismic facies. The study aims to reduce the computational time complexity of various existing unsupervised facies classification techniques. Further, to demonstrate the applicability, the proposed parallel frameworks are applied to the seismic dataset of Dutch Sector (North Sea) and the classification results are compared in terms of run-time complexity and accuracy. • Spark Based Framework to capture Facies present in Seismic data is proposed. • Proposed approach supports reduced runtime complexity while retaining desired accuracy. • Execution time comparison is carried out with state-of-the-art MPI based Implementation. • Performance results show the effectiveness of the Proposed Approach vs MPI based Implementation. [ABSTRACT FROM AUTHOR]

In the current scenario, exploring new means to gain accurate information regarding agriculture-related problems is the need of the hour. In this direction, we propose a multi-stage framework to perform spatial mapping and time series analysis on more than 26 million farmers' helpline call-log records, made available by the Ministry of Agriculture & Farmers' Welfare, Government of India. The proposed spatial analysis framework delivers hidden patterns regarding the crop-wise density of farmers calling for help from various regions of the country. Furthermore, the proposed step-plot concept gives insights into the time span of the problems in the agriculture sector. Additionally, the proposed framework explores the potential of high-end forecasting models, including five Deep Learning-based models to predict the topic-wise demand for help (number of query calls) by the producers of the target regions. To elaborate on the utility of the presented work, the article outlines two case studies corresponding to policy recommendations regarding agriculture extension and other related domains using AgriMine. • Spatial patterns regarding the Indian farmers' questions concerning mushroom crop. • Temporal insights regarding the farmers' telephonic queries concerning the rice crop. • Topic-wise planning of agricultural-extension activities based on past years' data. • TCN-based time-series forecasting of farmers' demand for telephonic help. • Performance comparison of Deep learning-based forecasting models'. [ABSTRACT FROM AUTHOR]

• Utilizing farmers' helpline data to forecast the number of query-calls. • Useful for agricultural policymaking and agriculture-based market analysis. • Telephonic records used to formulate time-series forecasting problem. • Eighty forecasting models built using four machine learning/deep learning methods. • Five different case studies to compare the models' performances. Forecasting the nationwide demand for agriculture-related help plays a crucial role in supporting the decision-making activities of the agri-sector. In this direction, the article presents an artificial intelligence-based framework for predicting the agriculture-related query-calls count in the nation's farmers-helpline network. The present work utilizes advanced data mining techniques to operate on the available dataset. The dataset utilized for the study includes over 1.3 million query-call logs accumulated from the "Kisan Call Center", a farmers' helpline network administered by the Ministry of Agriculture, Government of India. Moreover, to validate the proposed framework, we process data corresponding to the top-five rice-producing states of India. In addition, the study compares the forecasting performance of four Machine Learning and Deep Learning-based models, i.e., Support Vector Regression, Multi-layer Perceptron, Long Short-Term Memory Networks, and Gated Recurrent Units. The experimental results convey that the proposed framework is useful for predicting trends in farmers' problems. Furthermore, the framework is valuable in developing fully automated AI-based systems connected with the data servers of the Kisan Call Centers and providing the forecast in a mechanized manner. [ABSTRACT FROM AUTHOR]

Deep learning models address air quality forecasting problems far more effectively and efficiently than the traditional machine learning models. Specifically, Long Short-Term Memory networks (LSTMs) constitute a significant breakthrough in understanding the complex sequential behavioral dependencies of the time series. Further, LSTM models justify well with the speed–accuracy tradeoff, among other deep learning models. However, there are several limitations of such deep learning models. Firstly, the addition of multiple hidden layers, on the one hand, improves the performance but, on the other hand, requires extensive hardware and computation capabilities. Secondly, most of the previous works that utilized LSTMs for air quality forecasting do not consider the issue of optimal hyperparameter calibration. While deciding the gradient, network learning parameters should be so fixed such that the model does not underfit or overfit. To address these issues, a stochastic optimization algorithm, mimicking the pattern of flocking birds, is utilized to find the most fitting solution in the parameter search space. Particle swarm optimization setup primarily models varying particles representing parameters to reach an optimum state. Furthermore, the Spatio-temporal instabilities of LSTM models are addressed in this work using preprocessing, segmentation and feature engineering to understand seasonal and trend characteristics along with the Spatio-temporal correlation of the time series. The proposed model is employed on the air quality dataset of 15 locations in India. A variety of experiments are performed to prove the superiority of the proposed method. Firstly, a comparison with traditional sequential models and deep learning models is done. Secondly, results are further evaluated over several existing benchmark dataset samples. Results suggest that the proposed method outperforms existing forecasting models when evaluated over a variety of performance metrics. • Complex spatio-temporal dependencies in the data renders model learning task difficult. • Long Short-Term Memory network are utilized to model complex sequential dependencies. • Optimal hyperparameters selection using swarm intelligence is done. [ABSTRACT FROM AUTHOR]

The growing human population in this modern society hugely depends on the energy to fulfill their day-to-day needs and activities. Renewable energy sources, especially solar energy, can satisfy the global power demand while reducing global warming caused by conventional sources. Solar irradiance is an essential component in solar power applications. The availability of solar irradiance is influenced by several factors, such as forecasting horizon, weather classification, and performance evaluation metrics, which also need consideration. The accurate forecasting of solar irradiance is of utmost importance for the power system designers and grid operators for efficient management of solar energy systems. The intermittent and non-stationary nature of solar irradiance makes many existing statistical and machine learning approaches less competent in providing accurate predictions. In this context, deep learning models have been proposed by several researchers to reduce the limitations of existing machine learning models and improve prediction accuracy. In this work, an extensive and comprehensive review of deep learning-based solar irradiance forecasting models is presented. The effectiveness and efficacy of several deep learning models, including long short-term memory, deep belief network, echo state network, convolution neural network, etc. have been reviewed. The results obtained in the reported studies proved the superiority of deep learning models in solar forecasting applications. Few researchers have proposed deep hybrid models to improve the prediction performance further. A study reported that the hybrid of CNN-LSTM can enhance the prediction accuracy by 3.62%, 25.29%, 34.66%, 37.37% and 26.20% over CNN, LSTM, GRU, RNN and DNN, respectively. Overall, this paper offers preliminary guidelines for a detailed view of deep learning techniques that researchers and engineers can use to improve the solar photovoltaic plant's modeling and planning. [Display omitted] • A review of deep learning solar irradiance forecasting models is presented. • Factors that influence the accuracy of forecast models are also studied. • A comparative analysis of several deep learning models is conducted. • Hybrid deep learning models performed well over basic deep learning models. • Hybrid deep learning models are recommended for enhanced performance. [ABSTRACT FROM AUTHOR]

Research in text mining has recently gained a lot of importance due to the large increase in the number of electronic news articles, books, research papers, and e-mail messages. Clustering organizes text documents in an unsupervised fashion. In this paper, we propose an algorithm for clustering unstructured text documents using shape pattern matching. The Vector Space Model is used to represent our dataset as a term-weight matrix. The high dimensional vector space has been mapped to a two-dimensional plane that has the term weights plotted against a time axis. In this way, the text documents are represented in the form of time sequences. Initially, the documents are broadly grouped into categories that are determined using domain knowledge. The relevant portion of the document vector is then clipped out. The shape patterns present in these clipped portions are observed. Indexing of these shape patterns is done by preparing their alphabet. Grouping documents within a category which share the same shape pattern results in the required clusters. [ABSTRACT FROM AUTHOR]

Many scientific and business domains require the collection and analysis of time series data. Feature extraction is an important component of time series data mining. In this paper, we introduce simple and novel techniques for feature extraction from time series data based on moments and slopes. The proposed techniques are capable of handling vertical and horizontal shifts existing between time sequences. They can also handle global scaling and shrinking of the time sequences. [ABSTRACT FROM AUTHOR]

The volatile behavior of solar energy is the biggest challenge in its successful integration with existing grid systems. Accurate global horizontal irradiance (GHI) forecasting can resolve this issue and lead to early and effective participation in the energy market. This study proposes a new hybrid deep learning model, namely long short term memory–convolutional neural network (LSTM–CNN), for hourly GHI forecasting, which models the spatio-temporal features by integrating the long short term memory (LSTM) and convolutional neural network (CNN) model. The proposed model is trained with the meteorological data of 23 locations of California State, USA, which includes temperature, precipitation, relative humidity, cloud cover, etc., as input parameters. The proposed hybrid LSTM–CNN model firstly uses LSTM to extract the temporal features from time-series solar irradiance data, followed by CNN, which extracts the spatial features from the correlation matrix of several meteorological variables of target and its neighbor location. The prediction accuracy of the developed model is analyzed rigorously by examining the performance for a year, for four seasons and under three sky conditions. Besides, the proposed LSTM–CNN model shows a forecast skill score in a range of about 37%–45% over few standalone models, including smart persistence, support vector machine, artificial neural network, LSTM, CNN and other hybrid models. The findings of the present work suggest that the proposed hybrid LSTM–CNN model is a reliable alternative for short-term GHI prediction due to its high predictive accuracy under diverse climatic, seasonal and sky conditions. • A hybrid deep learning model, namely LSTM–CNN for hourly GHI forecasting is proposed. • LSTM–CNN extracts the spatio-temporal features from the dataset. • Model is trained with meteorological data of 23 locations of California State, USA. • Performance is evaluated for a year, four seasons and under three sky conditions. • The proposed model shows forecast skill score in a range of about 37%–45%. [ABSTRACT FROM AUTHOR]

The data article describes the Road Damage Dataset, RDD2022, encompassing of 47,420 road images from majorly six countries, Japan, India, the Czech Republic, Norway, the United States, and China. The dataset incorporates over 55,000 instances of road damage, specifically longitudinal cracks, transverse cracks, alligator cracks, and potholes. Designed to facilitate the development of deep learning methodologies for automated road damage detection and classification, RDD2022 was unveiled as part of the Crowd sensing‐based Road Damage Detection Challenge (CRDDC'2022), with a major contribution from the challenge winners. This challenge garnered global participation, urging researchers to propose solutions for automatic road damage detection in multiple countries. A noteworthy outcome of CRDDC'2022 was the emergence of a top‐performing model achieving a remarkable F1 Score of 76.9% for road damage detection in all six countries using RDD2022. This success underscores the dataset's practical applicability for municipalities and road agencies, enabling low‐cost, automatic monitoring of road conditions. Beyond its immediate utility, RDD2022 stands as a valuable benchmark for researchers in computer vision, geoscience, and machine learning, offering a rich resource for algorithmic evaluation in diverse image‐based applications, including classification and object detection. The latest big data cup, Optimized Road Damage Detection Challenge (ORDDC'2024), is also based on RDD2022, underscoring its continued relevance and pivotal role in current research and development endeavors. [ABSTRACT FROM AUTHOR]

India, the third largest electricity producer, has been suffering from high power deficit for the past few decades. In recent years, the Government of India (GOI) has taken several initiatives to provide uninterrupted energy availability to citizens. The proposed study is aimed to help Indian government and private agencies in proactive planning of various energy related activities including spatial resolved generation planning, timestamp activities planning, transmission network optimization and many more. In this context, the article delivers details about the performed Spatio-temporal analysis of energy availability/deficiency in the country, India. Furthermore, as a step towards overall electrification, the Indian government had launched the Saubhagya scheme (2017) to provide reliable and accelerated electricity access to un-electrified rural and urban households. The analytical assessment of citizens' views on such policies is beneficial for the government agencies to decide future spatial movement and resources planning. Hence, this study, later on, delivers a novel insight in the energy domain by examining citizens' views through Twitter-based regional analysis. This analysis gives inference regarding citizens' sentiments on policies or electrification status published by GOI. Here, the presented work showed that citizens from four out of the five regions express positive opinions on electrification progress made by GOI. In the last stage, the impact of increasing electricity demand and climate changes on air pollution of a city is examined using the Granger causality test. The test experiments are performed on the well-planned northern city, Chandigarh. Although many researchers have explored the association between the climatic conditions and the energy demand, in this work, along with these factors, the causal interaction with air pollution is also shown. • Spatio-temporal analysis of demand availability/deficiency is performed for country India. • A twitter-based regional analysis of citizens' opinions on electrification policies and issues is performed. • Causality interactions among changing electricity demand, climatic conditions and air pollution is well investigated. • Based on the experimental results, some useful insights for the policymakers are presented. [ABSTRACT FROM AUTHOR]

Prediction of solar irradiance is an essential requirement for reliable planning and efficient designing of solar energy systems. Thus, in present work, a new ensemble model, which consists of two advance base models, namely extreme gradient boosting forest and deep neural networks (XGBF-DNN), is proposed for hourly global horizontal irradiance forecast. These base models are integrated using ridge regression to avoid the over-fitting problem. Moreover, the proposed framework is designed carefully to ensure the diversity in base models, as diversity among base models is widely recognized as a key to the success of an ensemble model. Further, to enhance the model's accuracy, a subset of input features is selected, which includes temperature, clear-sky index, relative humidity, and hour of the day as the most relevant features. To provide a comprehensive and reliable assessment, proposed model is validated with data from three different climatic Indian locations. Subsequently, a seasonal analysis is also carried out to provide a deeper insight into the model's performance. The performance of the proposed model is evaluated by comparing the prediction results with different models, including benchmark smart persistence and traditional machine learning techniques, such as random forest, support vector regression, extreme gradient boosting forest and deep neural networks. The proposed ensemble model exhibits the best combination of stability and prediction accuracy irrespective of seasonal variations in weather conditions and shows a forecast skill score in a range of about 33%–40% in prediction error. The predictive performance and stability make XGBF-DNN an ideal and reliable model for hourly global horizontal irradiance prediction. Therefore, the developed model can be advised to serve as a potential prediction model in several domains, such as forecasting of solar power, wind power, electricity consumption, etc. [ABSTRACT FROM AUTHOR]

• Data mining based framework for mining association rules from farmers' helpline records. • Extracting hidden insights regarding the co-occurring agricultural problems of India. • Multi-criteria decision-making based system for selecting the most relevant insights. • Total 84 experiments are performed to investigate various characteristics of the proposed framework. • Framework for agricultural policymakers, prediction models, recommender systems, etc. Agricultural policymakers use various types of expert systems to identify agricultural problems and to explore their potential solutions. However, in the current scenario, there is no robust system that can be used to collect and analyze information regarding the problems faced by farmers of the developing countries on a large scale. This article outlines the possible mechanisms through which information and communication technology (ICT) with the use of Knowledge Discovery in Databases could facilitate agricultural adoption. The goal of this study is to explore data from a farmers' helpline center as a new medium to gain hidden insights in terms of association rules regarding the problems faced by Indian farmers. The dataset used in this study is collected from the "Kisan Call Center", a farmers' helpline center managed by the Ministry of Agriculture, Government of India. For this objective, we propose a new approach that uses association rule mining integrated with a multi-criteria decision-making technique, TOPSIS to extract only the most relevant patterns from the dataset. Later, we perform experiments in order to analyze the output of the proposed framework and verify the discovered knowledge against the validation data. The best experiment generates a rule-set, consisting of 702 association rules, including insights from 25 states of India, with an average confidence value of 73.21% on the validation data. The extracted inference reveals many hidden patterns regarding associations among the farmers' issues from the remote states of India. Finally, we identify various potential applications of our work and conclude with some possible future developments in the proposed approach. [ABSTRACT FROM AUTHOR]

The present work focuses on predicting users’ next place of visit using their past tweets. We hypothesize that tweets of the person have predictive power on his location and therefore can be used to predict his next place of visit. This problem is important for location based advertising and recommender based services. To predict the next place of visit, we calculate the probabilities of visiting different types of places using bank of binary classifiers and Markov models. More specifically, we train bank of binary classifiers on past tweets and calculated the probabilities of visiting next places. Since bank of binary classifiers is based on a bag-of-words model, to account for time of last visited place and place itself, we built Markov models for different time duration to calculate probabilities of visiting next place. Empirical evaluation shows that by combining the probabilities obtained from bank of binary classifiers and Markov models the accuracy of predicting next place increased from 65% to 80%. [ABSTRACT FROM AUTHOR]

Privacy preservation in distributed database is an active area of research. With the advancement of technology, massive amounts of data are continuously being collected and stored in distributed database applications. Indeed, temporal associations and correlations among items in large transactional datasets of distributed database can help in many business decision-making processes. One among them is mining frequent itemset and computing their association rules, which is a nontrivial issue. In a typical situation, multiple parties may wish to collaborate for extracting interesting global information such as frequent association, without revealing their respective data to each other. This may be particularly useful in applications such as retail market basket analysis, medical research, academic, etc. In the proposed work, we aim to find frequent items and to develop a global association rules model based on the genetic algorithm (GA). The GA is used due to its inherent features like robustness with respect to local maxima/minima and domain-independent nature for large space search technique to find exact or approximate solutions for optimization and search problems. For privacy preservation of the data, the concept of trusted third party with two offsets has been used. The data are first anonymized at local party end, and then, the aggregation and global association is done by the trusted third party. The proposed algorithms address various types of partitions such as horizontal, vertical, and arbitrary. [ABSTRACT FROM AUTHOR]

The article discusses an accurate and valid model developed for predicting the survivability of burn patients. Topics covered include the different classification algorithms used in developing the proposed prediction model, the methods and performance measures used in comparing the classification algorithms, and the analyzed experimental results obtained using real medical data from Swami Ramanand Tirth (SRT) Hospital in India.

Abstract: A wide range of computational methods and tools for data analysis are available. In this study we took advantage of those available technological advancements to develop prediction models for the prediction of a Type-2 Diabetic Patient. We aim to investigate how the diabetes incidents are affected by patients’ characteristics and measurements. Efficient predictive modeling is required for medical researchers and practitioners. This study proposes Hybrid Prediction Model (HPM) which uses Simple K-means clustering algorithm aimed at validating chosen class label of given data (incorrectly classified instances are removed, i.e. pattern extracted from original data) and subsequently applying the classification algorithm to the result set. C4.5 algorithm is used to build the final classifier model by using the k-fold cross-validation method. The Pima Indians diabetes data was obtained from the University of California at Irvine (UCI) machine learning repository datasets. A wide range of different classification methods have been applied previously by various researchers in order to find the best performing algorithm on this dataset. The accuracies achieved have been in the range of 59.4–84.05%. However the proposed HPM obtained a classification accuracy of 92.38%. In order to evaluate the performance of the proposed method, sensitivity and specificity performance measures that are used commonly in medical classification studies were used. [ABSTRACT FROM AUTHOR]

• The automatic monitoring of road conditions for multiple countries is addressed. • Deep Learning models are trained for detecting road damages in India, Japan, and Czech. • Recommendations are provided for reusing the data and models released by any country. • A large-scale road damage dataset comprising 26,620 annotated road images is proposed. • The Global Road Damage Detection Challenge'2020 utilizes a part of the proposed data. Many municipalities and road authorities seek to implement automated evaluation of road damage. However, they often lack technology, know-how, and funds to afford state-of-the-art equipment for data collection and analysis of road damages. Although some countries, like Japan, have developed less expensive and readily available Smartphone-based methods for automatic road condition monitoring, other countries still struggle to find efficient solutions. This work makes the following contributions in this context. Firstly, it assesses usability of Japanese model for other countries. Secondly, it proposes a large-scale heterogeneous road damage dataset comprising 26,620 images collected from multiple countries (India, Japan, and the Czech Republic) using smartphones. Thirdly, it proposes models capable of detecting and classifying road damages in more than one country. Lastly, the study provides recommendations for readers, local agencies, and municipalities of other countries when one other country publishes its data and model for automatic road damage detection and classification. A part of the proposed dataset was utilized for Global Road Damage Detection Challenge'2020 and can be accessed at (https://github.com/sekilab/RoadDamageDetector/). [ABSTRACT FROM AUTHOR]