Article
Psychology, Multidisciplinary
Anmar Abdul-Rahman
Summary: The Heisenberg-Gabor uncertainty principle defines the limits of information resolution in both time and frequency domains, while wavelet transformation provides a workable compromise by decomposing the signal in both time and frequency. This study aimed to compare the accuracy of predictive models in mental arithmetic in time and frequency domains.
FRONTIERS IN PSYCHOLOGY
(2022)
Article
Computer Science, Artificial Intelligence
Ranjit Kumar Paul, Sandip Garai
Summary: Accurate forecasting in Indian agriculture is crucial, with machine learning techniques like artificial neural network and wavelet transformation being effective in handling nonlinear datasets to improve model accuracy.
Article
Mathematics
Nurul Aityqah Yaacob, Jamil J. Jaber, Dharini Pathmanathan, Sadam Alwadi, Ibrahim Mohamed
Summary: This study uses various discrete wavelet transform filters to model and forecast the time-dependent mortality index of the Lee-Carter model, with the MODWT-ARIMA (5,1,0) model with the BL14 filter showing the best fit for the log of death rates data across five countries. Using MODWT leads to improvements in forecasting mortality rates within the standard framework of the LC model.
Article
Environmental Sciences
Erdinc Aladag
Summary: This study successfully predicted the future PM10 concentration in Erzurum, Turkey using a hybrid WT-ARIMA model, which was more effective than the traditional ARIMA model. The developed model can serve as a reference for particulate matter pollution early warning in regions with high air pollution levels.
Article
Mathematics, Interdisciplinary Applications
Madhurima Panja, Tanujit Chakraborty, Sk Shahid Nadim, Indrajit Ghosh, Uttam Kumar, Nan Liu
Summary: Dengue fever is a widespread virulent disease that affects millions of people globally and puts a strain on healthcare systems. Due to the lack of specific drugs and vaccines, policymakers rely on early warning systems for intervention decisions. However, existing forecasting models often provide unstable and unreliable forecasts. This study proposes a new model called XEWNet that incorporates wavelet transformation into an ensemble neural network framework to improve the accuracy and reliability of dengue outbreak predictions.
CHAOS SOLITONS & FRACTALS
(2023)
Article
Engineering, Electrical & Electronic
Akshita Gupta, Arun Kumar, K. Boopathi
Summary: The rapid expansion of Renewable Energy Sources (RES) worldwide, particularly solar and wind energy, has led to concerns regarding their intermittency and unpredictability at the grid level. This study focuses on intraday wind power forecasting using the Wavelet-ARIMA model to improve accuracy and reliability.
ELECTRIC POWER SYSTEMS RESEARCH
(2021)
Article
Computer Science, Artificial Intelligence
Madhurima Panja, Tanujit Chakraborty, Uttam Kumar, Nan Liu
Summary: Infectious diseases continue to be a major cause of illness and death worldwide, with epidemic waves of infection. The lack of specific drugs and vaccines exacerbates the situation, leading to a reliance on accurate and reliable epidemic forecasters. The proposed Ensemble Wavelet Neural Network (EWNet) model effectively characterizes the non-stationary behavior and seasonal dependencies of epidemic time series, improving the accuracy of epidemic forecasting compared to other methods.
Article
Agronomy
Santosha Rathod, Amit Saha, Rahul Patil, Gabrijel Ondrasek, Channappa Gireesh, Madhyavenkatapura Siddaiah Anantha, Dhumannatarao Venkata Krishna Nageswara Rao, Nirmala Bandumula, Ponnuvel Senguttuvel, Arun Kumar Swarnaraj, Shaik N. Meera, Amtul Waris, Ponnuraj Jeyakumar, Brajendra Parmar, Pitchiahpillai Muthuraman, Raman Meenakshi Sundaram
Summary: Accurate rice yield forecasting is crucial for cereal production planning and decision-making. A new two-stage STARMA approach was developed to improve prediction accuracy in intensive national rice agroecosystems, offering a promising alternative for estimating crop yields and patterns in agricultural systems.
Article
Mathematics
Shuqi Wang, Huajun Zhang, Xuetao Zhang, Yixin Su, Zhenghua Wang
Summary: This paper proposes a speaker recognition method based on the PWPE-ECA-Res2Net-TDNN model, which demonstrates strong robustness and noise resistance in cross-scenario applications, and maintains a relatively short recognition time even under the highest recognition rate conditions. A set of ablation experiments targeting each module of the proposed model indicate that each module contributes to an improvement in the recognition performance.
Article
Water Resources
Reza Rezaiy, Ani Shabri
Summary: Climate change and water supply shortage are critical global issues. Early forecasting of drought is crucial for strategic planning and water supply management. This study presents a hybrid W-ARIMA model for drought forecasting, which outperforms the traditional ARIMA model based on the SPI index. The empirical results demonstrate the superior accuracy of the proposed W-ARIMA model for drought prediction.
JOURNAL OF WATER AND CLIMATE CHANGE
(2023)
Article
Computer Science, Information Systems
Khudhayr A. Rashedi, Mohd Tahir Ismail, Abdeslam Serroukh, S. Al Wadi
Summary: We introduce a new wavelet based procedure for detecting outliers in financial discrete time series. The procedure focuses on analyzing residuals obtained from a model fit, and is applicable to various models, including the Generalized Autoregressive Conditional Heteroskedasticity (GARCH) model. Our methodology has several advantages over existing methods, such as not requiring the series sample size to be a power of 2 and being able to explore any wavelet filter. The proposed method's efficiency is demonstrated through its application to real series.
CMC-COMPUTERS MATERIALS & CONTINUA
(2022)
Article
Engineering, Marine
Shuguang Zhang, Jijian Lian, Jinxuan Li, Fang Liu, Bin Ma
Summary: This study investigates the nonlinear characteristics of waves generated by submerged jets through experimental measurements. The results show that the amplitude of wave surface varies with the variation of measuring point position and upstream water depth, and the wave surface tilts forward to backward along the propagation path. The energy of waves increases with the increase of upstream water depth, and the energy transfer caused by wave interaction is weak according to the wavelet-based bicoherence distribution.
Article
Statistics & Probability
Xin Zhao, Stuart Barber, Charles C. Taylor, Zoka Milan
Summary: The study proposes two methods based on regression trees: one ensemble method and one method based on a single tree, which are suitable for forecasting and updating interval predictions. According to the results of the experiments, both methods perform well, with high coverage and relatively quick recovery after changes in the data.
ADVANCES IN DATA ANALYSIS AND CLASSIFICATION
(2021)
Article
Ecology
Sareh Hashem Geloogerdi, Abbasali Vali, Mohammad Reza Sharifi
Summary: Desertification is a critical global environmental issue, particularly in arid and semiarid regions. This study uses a feature space model and time series models to simulate desertification trends. By calculating remote sensing indexes and applying breakpoint classification, different levels of desertification are identified. The most accurate models are selected based on the analysis of autocorrelation and partial autocorrelation functions.
JOURNAL OF ARID ENVIRONMENTS
(2023)
Article
Chemistry, Analytical
Alina Barbulescu, Cristian Stefan Dumitriu
Summary: Experiments have demonstrated the presence of electrical signals in the ultrasonic cavitation field, and their properties have been studied. This research models the voltage collected in seawater during ultrasonic cavitation produced by a 20kHz frequency generator and compares the effectiveness of different modeling methods, with the hybrid Wavelet-ANN model proving to be the most accurate.
Article
Environmental Sciences
B. B. Gogoi, A. Borgohain, K. Konwar, J. G. Handique, R. K. Paul, P. Khare, H. Malakar, J. Saikia, T. Karak
Summary: The study reveals that National Highway can have damaging effect on soil chemical properties, leading to higher levels of soluble salts, organic carbon, total nitrogen, available phosphorous, potassium, and metal concentrations such as copper, iron, manganese, and zinc in top soils compared to sub soils near the highway. The application of a generalized linear model shows significant differences in metal content across various sampling distances from the highway. The study also highlights the accumulation of traffic-related metals in top soils near the National Highway in tea-growing regions.
INTERNATIONAL JOURNAL OF ENVIRONMENTAL SCIENCE AND TECHNOLOGY
(2022)
Article
Agronomy
Ranjit Kumar Paul, Tanmoy Karak
Summary: This paper conducts horizontal and vertical integration on the wholesale and retail prices of wheat in major markets in India. The findings show that price signals are transmitted across regions, and there is significant presence of asymmetric and nonlinear cointegration.
Article
Chemistry, Applied
Arup Borgohain, Mridusmita Sarmah, Kaberijyoti Konwar, Rimjim Gogoi, Bidyot Bikash Gogoi, Puja Khare, Ranjit Kumar Paul, Jyotirekha G. Handique, Harisadhan Malakar, Diganta Deka, Jiban Saikia, Tanmoy Karak
Summary: The use of tea pruning litter biochar significantly reduces the concentration of arsenic, cadmium, and chromium in tea leaves, without adverse health effects on tea consumers.
Article
Multidisciplinary Sciences
Ranjit Kumar Paul, Md. Yeasin, Pramod Kumar, Prabhakar Kumar, M. Balasubramanian, H. S. Roy, A. K. Paul, Ajit Gupta
Summary: This study explores the efficient use of machine learning algorithms, such as GRNN, SVR, RF, and GBM, to forecast the wholesale price of Brinjal in seventeen major markets in Odisha, India. The results show that GRNN performs better compared to other techniques in most cases.
Article
Environmental Sciences
Md Yeasin, Dipanwita Haldar, Suresh Kumar, Ranjit Kumar Paul, Sonaka Ghosh
Summary: This study examines the potential of Sentinel-1 and Sentinel-2 satellites for monitoring sugarcane phenology and finds that their combination is more efficient. Key features include backscatters and parameters from Sentinel-1, and vegetation indices from Sentinel-2.
Article
Agronomy
Ranjit Kumar Paul, Sengottaiyan Vennila, Md Yeasin, Satish Kumar Yadav, Shabistana Nisar, Amrit Kumar Paul, Ajit Gupta, Seetalam Malathi, Mudigulam Karanam Jyosthna, Zadda Kavitha, Srinivasa Rao Mathukumalli, Mathyam Prabhakar
Summary: This study examined the influence of weather variables on the occurrence of spiders in pigeon pea across different agro-climatic zones in India. It also developed forecast models and compared their performance. The findings showed that the wavelet-ANN model had better prediction accuracy compared to other models. The study also proposed using the model in conjunction with pest-defender ratios to reduce insecticidal sprays and add ecological and economic value to the integrated pest management of pigeon pea insects.
Article
Engineering, Environmental
Mridusmita Sarmah, Arup Borgohain, Bidyot Bikash Gogoi, Md Yeasin, Ranjit K. Paul, Harisadhan Malakar, Jyotirekha G. Handique, Jiban Saikia, Diganta Deka, Puja Khare, Tanmoy Karak
Summary: This study investigated the effects of tea pruning litter biochar (TPLBC) on the accumulation of copper (Cu), manganese (Mn), and zinc (Zn) in soil, their uptake by tea plants, and the level of soil contamination. The results showed that the application of TPLBC did not have any adverse effect on soil. The best treatment for Cu, Mn, and Zn was the application of 400 kg TPLBC ha-1.
JOURNAL OF HAZARDOUS MATERIALS
(2023)
Article
Multidisciplinary Sciences
Ranjit Kumar Paul, Tanima Das, Md Yeasin
Summary: Forecasting price volatility of agricultural commodities is of great importance nowadays. Traditional parametric models are inefficient in capturing volatility in price series. Therefore, machine learning techniques such as support vector regression (SVR) may be applied to improve forecasting accuracy. This study proposes an algorithm based on a combination of the generalized autoregressive conditional heteroscedastic (GARCH) model and supervised machine learning (e.g. SVR) for forecasting onion price volatility in two major markets in India, Delhi and Kolkata. Empirical results demonstrate the outperformance of the proposed algorithm compared to the GARCH model using metrics such as Root Mean Square Error, Mean Absolute Error, and R-2 log.
NATIONAL ACADEMY SCIENCE LETTERS-INDIA
(2023)
Article
Genetics & Heredity
Jutan Das, Sanjeev Kumar, Dwijesh Chandra Mishra, Krishna Kumar Chaturvedi, Ranjit Kumar Paul, Amit Kairi
Summary: CRISPR-Cas9 system is a widely used genome editing technique, but it has off-target effects. In this paper, machine learning algorithms were used to develop models for predicting CRISPR-Cas9 cleavage sites. The models based on random forest technique performed the best, with an accuracy of 96.27% and an AUC value of 99.21%. Among the artificial neural network and support vector machine models, ANN1-ReLU and SVM-Linear showed better performance.
FRONTIERS IN GENETICS
(2023)
Article
Genetics & Heredity
Amrit Kumar Paul, Himadri Shekhar Roy, Ranjit Kumar Paul, Prakash Kumar, Md Yeasin
Summary: In the fields of plant and animal breeding, it is common to have correlated observations which violate the assumption of independence. This study focuses on the estimation of heritability in the presence of correlated errors, specifically considering autoregressive error structures such as AR(1) and AR(2) models, which have a significant impact on heritability estimates.
Article
Mathematics
Debopam Rakshit, Ranjit Kumar Paul, Md Yeasin, Walid Emam, Yusra Tashkandy, Christophe Chesneau
Summary: Seasonal production, weather abnormalities, and perishability contribute to the volatility of potato prices. Asymmetric price volatility occurs when positive and negative shocks have different effects. GARCH is a symmetric model that cannot capture this asymmetry, while EGARCH, APARCH, and GJR-GARCH models are popular for capturing asymmetric price volatility. This paper attempts to model the price volatility of potatoes in various markets using these models and confirms the presence of asymmetry through News Impact Curves (NICs).
Article
Agriculture, Multidisciplinary
Surajit Mondal, Debashis Chakraborty, Ranjit Kumar Paul, Arun Mondal, J. K. Ladha
Summary: No-till is considered more as a climate change mitigation option than a practice for managing soil organic C content. A global meta-analysis was conducted, showing that no-till significantly increased SOC concentration by 38% in the 0-5 cm soil layer compared to conventional tillage.
AGRICULTURE ECOSYSTEMS & ENVIRONMENT
(2023)
Article
Mathematics
Sandip Garai, Ranjit Kumar Paul, Debopam Rakshit, Md Yeasin, Walid Emam, Yusra Tashkandy, Christophe Chesneau
Summary: In this study, different wavelet filters were used to predict agricultural commodity prices, and the combination of Haar filter with random forest model performed the best in terms of prediction accuracy.
Article
Engineering, Environmental
Md Yeasin, Ranjit Kumar Paul, Sampa Das, Diganta Dek, Tanmoy Karak
Summary: The onset of the COVID-19 pandemic led to a nationwide lockdown in India, resulting in a decline in air pollution. Data analysis showed a significant impact on air quality parameters during the lockdown period. Insights from this pandemic will be valuable in improving ambient air quality while considering economic growth.
JOURNAL OF HAZARDOUS MATERIALS ADVANCES
(2023)
Article
Multidisciplinary Sciences
Ranjit Kumar Paul, Md Yeasin, Pramod Kumar, A. K. Paul, H. S. Roy
Summary: Vegetable prices are difficult to forecast due to various factors, such as weather, demand and supply chain, and government policies, resulting in volatile fluctuations. This study compares traditional statistical models, machine learning, and deep learning techniques to find the best-suited model for accurate and timely vegetable forecast. Using cauliflower markets as an empirical illustration, machine learning and deep learning techniques are applied to identify price complexity. The deep learning model outperforms traditional time-series models and machine learning techniques for both short- and long-term forecasting.