Supervised Classification Techniques Research Articles

Coupling of Rough Set Theory and Predictive Power of SVM Towards Mining of Missing Data

Rough set theory offers a novel approach to identifying structural correlations amidst imprecise or noisy data, particularly applicable to variables with diverse values. It presents a promising avenue for handling fuzzy, conflicting, and uncertain data, with recent models incorporating various fuzzy generalizations. This technique stands out as a popular solution within artificial intelligence, particularly in data analysis and processing tasks. In the medical domain, where missing data poses a significant challenge, leveraging rough set theory alongside machine learning algorithms for disease prediction is common. This paper proposes a model that effectively predicts missing values using rough set theory, addressing the prevalent issue of incomplete data. By providing a systematic approach and robust algorithm, the model demonstrates the adaptability and potential of rough set theory in contemporary data analysis scenarios. Classification of the predicted data set using supervised Learning Model (SVM) results in accuracy of 82.1% while the F1 score is 82.6%. Through validation with reallife medical datasets using supervised classification techniques, the paper underscores the accuracy and applicability of the proposed algorithm, offering a valuable tool for researchers and practitioners grappling with the complexities of modern data analysis.

Determination of change in the land use and land cover of the Samsun Bafra Delta Plain from 1990 to 2020 using GIS and Remote Sensing Techniques

Land use and land cover changes can have detrimental effects on the ecology, if they are not properly aligned with the characteristics of the land. This study aims to evaluate the temporal changes in land use and land cover of Bafra Delta plain, situated in the east of Samsun province. The region is one of the most significant plains within the Black Sea area. Remote sensing technique was utilized in this research which made use of Landsat images from 1990, 2000, 2010, and 2020. Supervised classification was applied in ENVI 5.3v software to perform calculations, resulting in six main classes. Field work was applied to classify the unclassified classes. The resulting six land use-land cover classes were agriculture lands, forest, dune, marshy, water surface, and artificial areas. To determine land use efficiency, analogue data was digitised and transferred to a GIS database. The agricultural areas occupy the largest portion of the plain, followed by hazelnut and artificial areas. The changes over the last decade, notably the growth of artificial areas and water surfaces, and the reduction of arable lands, highlight significant variations in size across the areas. Furthermore, the study indicated that remote sensing and geographic information system techniques play a crucial role in identifying and monitoring land cover and land use trends on a large-scale to produce accurate and timely data. Poorly adapted land use changes can cause major ecological damage. The aim of this study is to identify the changes over time in land use and land cover of Bafra Delta plain, located to the east of Samsun city and one of the most significant plains in the Black Sea region, using remote sensing techniques. To this end, Landsat images from 1990, 2000, 2010 and 2020 are utilized. To perform the calculations, ENVI 5.3v software was employed, applying a supervised classification technique that resulted in forming six main classes. Fieldwork was conducted to classify the unclassified classes. The resulting land-use and land-cover classes were agricultural land, forest, dunes, marshland, water surface, and artificial areas. To evaluate land-use efficiency, analogue data were digitalised and imported into a GIS database. The plain's most extensive land-use areas consist of agricultural lands, followed by hazelnut and artificial areas. In the last decade, the rise in artificial and water surfaces and the decline in agricultural areas highlights significant changes in the region's size. This study also emphasises the crucial role of remote sensing and geographic information system techniques in generating fast and consistent data for monitoring large-scale land cover and land use trends.

Detecting, Analyzing, and Predicting Land Use/Land Cover (LULC) Changes in Arid Regions Using Landsat Images, CA-Markov Hybrid Model, and GIS Techniques

Understanding the change dynamics of land use and land cover (LULC) is critical for efficient ecological management modification and sustainable land-use planning. This work aimed to identify, simulate, and predict historical and future LULC changes in the Sohag Governorate, Egypt, as an arid region. In the present study, the detection of historical LULC change dynamics for time series 1984–2002, 2002–2013, and 2013–2022 was performed, as well as CA-Markov hybrid model was employed to project the future LULC trends for 2030, 2040, and 2050. Four Landsat images acquired by different sensors were used as spatial–temporal data sources for the study region, including TM for 1984, ETM+ for 2002, and OLI for 2013 and 2022. Furthermore, a supervised classification technique was implemented in the image classification process. All remote sensing data was processed and modeled using IDRISI 7.02 software. Four main LULC categories were recognized in the study region: urban areas, cultivated lands, desert lands, and water bodies. The precision of LULC categorization analysis was high, with Kappa coefficients above 0.7 and overall accuracy above 87.5% for all classifications. The results obtained from estimating LULC change in the period from 1984 to 2022 indicated that built-up areas expanded to cover 12.5% of the study area in 2022 instead of 5.5% in 1984. This urban sprawl occurred at the cost of reducing old farmlands in old towns and villages and building new settlements on bare lands. Furthermore, cultivated lands increased from 45.5% of the total area in 1984 to 60.7% in 2022 due to ongoing soil reclamation projects in desert areas outside the Nile Valley. Moreover, between 1984 and 2022, desert lands lost around half of their area, while water bodies gained a very slight increase. According to the simulation and projection of the future LULC trends for 2030, 2040, and 2050, similar trends to historical LULC changes were detected. These trends are represented by decreasing desert lands and increasing urban and cultivated newly reclaimed areas. Concerning CA-Markov model validation, Kappa indices ranged across actual and simulated maps from 0.84 to 0.93, suggesting that this model was reasonably excellent at projecting future LULC trends. Therefore, using the CA-Markov hybrid model as a prediction and modeling approach for future LULC trends provides a good vision for monitoring and reducing the negative impacts of LULC changes, supporting land use policy-makers, and developing land management.

Assessment of future prediction of urban growth and climate change in district Multan, Pakistan using CA-Markov method

This research utilized remote sensing (RS) technology to analyze and monitor land use land cover (LULC) and climate change. This work mainly involves three components: (i) investigating the historical trends of LULC and land surface temperature (LST) using Landsat data for 1990, 2005, and 2020. (ii)Evaluate the impact of climate change on urban temperature for increased heat stress in the Multan region and (iii) Future predictions of LULC and LST for 2035 and 2050 by the CA-Markov Chain model. Supervised classification techniques were employed to classify the images into various LULC classes. Historical trends of LULC revealed that over the study period, the built-up area in Multan has increased by almost 30,435 Ha (8.34%), which would continue to reach40003 ha (10.96%) from 1990 to 2020 at the cost of vegetation loss, which is countlessly decreasing and will get from 283,145 Ha (−77.57%) in 2035 to264443 Ha (−71.4%) till 2050. Further, the findings about LST revealed that over the region, the average LST values have increased from 1.2 °C over the last two decades and will additionally include 1.8 °C until 2050.This study underscores the importance of RS technology in understanding past LULC changes and predicting future land transformations and temperature variations. The results contribute valuable insights for policymakers, land managers, and urban planners to make informed decisions for sustainable LULC and climate adaptation strategies in the face of growing urbanization and climate change challenges.

Socioeconomic Drivers of Land Use and Land Cover Change in Western Ethiopia

A variety of socioeconomic and environmental drivers have contributed to changes in LULC around the world in recent years. This study examines the socioeconomic drivers that accelerated LULC in western Ethiopia. The data were generated from terrestrial satellite images primary and secondary sources. Primary data sources include household surveys, field observations, group discussions, interviews, key informants, and interpreting remote sensing data. Secondary data were reviewed mainly from relevant literature both published and unpublished materials. Landsat images were classified using the supervised classification technique and maximum likelihood classifier using arc GIS 10.3 to create LULC maps of the study area. Accuracy score and kappa coefficient were used to confirm the accuracy of the classified LULC, and agricultural land, settlement, bare land, forest land, and water body were the main LULC classes in the district. Forest cover in three decades (1990–2020) in the study area decreased from 12.1% in 1990 to 2.6% in 2020. The data were also analyzed using a descriptive model, Pearson correlation, and binary logistic regression. The independent variables (age and gender) show a Pearson’s positive correlation with the drivers of LULC dynamics; that is, as these independent variables increase, the drivers of LULC dynamics also increase, whereas educational status and land holding size show a negative correlation. This shows that the drivers of the anthropogenic forces of LULC dynamics decreased as the number of educated populations and the size of land holdings increased, and vice versa. Then, the binary logistic regression model examined the relationship between the dependent and the major socioeconomic (independent) variables. Logistic regression was performed to determine how independent variables and the drivers of LULC (natural forces or anthropogenic forces) change and the model was statistically significant (x2 = 23.971, df = 5, P < 0.001). The model explained 13.9% (Nagelkerke R2) of the variance in the drivers of LULC dynamics and correctly classified 66.1% of the cases. The study found that age, gender, and educational status largely determine the drivers of LULC dynamics and have the greatest chance of determining the anthropogenic forces. Therefore, relevant stakeholders should take integrated measures to reduce the drivers of LULC dynamics through landscape restoration.

Land use/land change detection and determination of land surface temperature variation in green belt (Nasirabad) district of Balochistan, Pakistan

The current study determined the changes in Land Use/Land Change (LU/LC) and variation in land surface temperature (LST) in the Green Belt (Nasirabad district) area of Balochistan, Pakistan. To achieve this, we used GIS software (ArcMap 10.7.1) to analyze remote sensing data acquired from Landsat imagery taken in 1993, 2003, 2013, and 2023. A supervised classification technique using the maximum likelihood algorithm (MLC) was applied to generate a ground-truth LU/LC classification. Based on our findings, almost 415.28 km2 (− 12.89%) of formerly undeveloped land has been transformed into urban neighborhoods and green spaces during the last three decades. Between 1993 and 2023, the study area gained 288.29 km2 (8.94%) of vegetation and 136.10 km2 (4.22%) of settled land. Minimum, maximum, and average LST changes were recorded as 7.50, − 5.06, and 1.22 °C for the whole thirty years. Overall, the analysis data showed that an increase in human settlements in the area investigated led to a rise in mean LST (1.22 °C). Finally, GIS and RS may be used together to track land usage over time, a crucial piece of data for eco-friendly planning. While the LU/LC and LST provide valuable insights into the rational and optimal use of land resources, the implications of policy remain constrained.

Machine learning based hybrid anomaly detection technique for automatic diagnosis of cardiovascular diseases using cardiac sympathetic nerve activity and electrocardiogram.

Coronary artery diseases (CADs) are the leading cause of death worldwide and early diagnosis is crucial for timely treatment. To address this, our study presents a novel automated Artificial Intelligence (AI)-based Hybrid Anomaly Detection (AIHAD) technique that combines various signal processing, feature extraction, supervised, and unsupervised machine learning methods. By jointly and simultaneously analyzing 12-lead cardiac sympathetic nerve activity (CSNA) and electrocardiogram (ECG) data, the automated AIHAD technique performs fast, early, and accurate diagnosis of CADs. In order to develop and evaluate the proposed automated AIHAD technique, we utilized the fully labeled STAFF III and PTBD databases, which contain the 12-lead wideband raw recordings non-invasively acquired from 260 subjects. Using these wideband raw recordings, we developed a signal processing technique that simultaneously detects the 12-lead CSNA and ECG signals of all subjects. Using the pre-processed 12-lead CSNA and ECG signals, we developed a time-domain feature extraction technique that extracts the statistical CSNA and ECG features critical for the reliable diagnosis of CADs. Using the extracted discriminative features, we developed a supervised classification technique based on Artificial Neural Networks (ANNs) that simultaneously detects anomalies in the 12-lead CSNA and ECG data. Furthermore, we developed an unsupervised clustering technique based on Gaussian mixture models (GMMs) and Neyman-Pearson criterion, which robustly detects outliers corresponding to CADs. Using the automated AIHAD technique, we have, for the first time, demonstrated a significant association between the increase in CSNA signals and anomalies in ECG signals during CADs. The AIHAD technique achieved highly reliable detection of CADs with a sensitivity of 98.48 %, specificity of 97.73 %, accuracy of 98.11 %, positive predictive value of 97.74 %, negative predictive value of 98.47 %, and F1-score of 98.11 %. Hence, the automated AIHAD technique demonstrates superior performance compared to the gold standard diagnostic test ECG in the diagnosis of CADs. Additionally, it outperforms other techniques developed in this study that separately utilize either only CSNA data or only ECG data. Therefore, it significantly increases the detection performance of CADs by taking advantage of the diversity in different data types and leveraging their strengths. Furthermore, its performance is comparatively better than that of most previously proposed machine and deep learning methods that exclusively used ECG data to diagnose or classify CADs. Additionally, it has a very low implementation time, which is highly desirable for real-time detection of CADs. The proposed automated AIHAD technique may serve as an efficient decision-support system to increase physicians' success in fast, early, and accurate diagnosis of CADs. It may be highly beneficial and valuable, particularly for asymptomatic patients, for whom the diagnostic information provided by ECG alone is not sufficient to reliably diagnose the disease. Hence, it may significantly improve patient outcomes by enabling timely treatments and considerably reducing the mortality of cardiovascular diseases (CVDs).

Detection of Land Use/Land Cover Change and Land Surface Temperature in the Eastern Part of Batna City (North East Algeria) Using Remote Sensing Data and GIS


 
 
 The present paper aims to evaluate the accuracy of classifying Land Use /Land Cover (LULC) types and assesses the trends of their changes in the Eastern Part of Batna City (Northernf Algeria) using remote sensing and GIS. The accuracy of image Land Satellite (Land Sat ) was evaluated using the supervised classification technique, it’s applied in multi spectral and multi temporal satellite data acquired in 2000,2010,2022 and assessed with GOOGLE EARTH PRO and IMAGERY Land Use and topographical map. The second part focused on extraction of LST in three phases and explored the relationship between two land cover indices (NDVI, NDBI) and LST.LU/LC detected, quantified, and stati- cally analyzed, the result indicate that from 2000-2022 the built-up areas increased by 0.34% (6.638km2), the forest area increased by 1.8% (35.144km2), agricultural land cover increased by 1.12% (21.867km2), while bare land decreased by 2.17% (42.368km2). The conversions of areas from bare land to urban land represent the most significant Land Cover changes. The accuracy assessment and correlation coefficient R2 analysis in this study affirms the previous research findings. Even a single land use unit like built-up area, bare land and vegetation also create differences in LST (R2 of NDBI vs. LST ranges from 0.64 to 0.79; NDVI vs. LST ranges from -0.73 to -0.82). With the change of the LU/LC style, its imprint is reflected on the LST. Therefore, immediate reflection on new urbanism must be adopted, initiated and implemented to stop the warming that contributes to climate change in the study area.
 
 

A Comparative Analysis of Supervised Classification Algorithms and Missing Data Handling for Enhancing Chronic Kidney Disease Prediction

Chronic kidney disease (CKD), which is becoming a more significant public health concern, is characterized by a gradual but concerning increase in morbidity and death, particularly in its early, asymptomatic stages. Risk factors for chronic kidney disease (CKD), including genetic predisposition, obesity, diabetes, and hypertension, affect the illness's prevalence. When there are no outward signs of an illness, it is challenging to diagnose and treat it in its early stages. To tackle this pressing issue, our research does a comprehensive investigation through a comparative comparison of supervised classification techniques. In particular, we examine the prediction performance of CKD using the Random Forest, Decision Tree, and Support Vector Machine (SVM) techniques. We also look into a number of approaches to handling missing data. Our research presents a thorough evaluation of these algorithms' performance under different data cleaning methods, pointing out both their benefits and drawbacks. Ultimately, our research aims to clarify the early detection and treatment of chronic kidney disease (CKD) and pave the way for larger-scale public health initiatives to tackle this quickly escalating health emergency.

Using PRISMA Hyperspectral Data for Land Cover Classification with Artificial Intelligence Support

Hyperspectral satellite missions, such as PRISMA of the Italian Space Agency (ASI), have opened up new research opportunities. Using PRISMA data in land cover classification has yet to be fully explored, and it is the main focus of this paper. Historically, the main purposes of remote sensing have been to identify land cover types, to detect changes, and to determine the vegetation status of forest canopies or agricultural crops. The ability to achieve these goals can be improved by increasing spectral resolution. At the same time, improved AI algorithms open up new classification possibilities. This paper compares three supervised classification techniques for agricultural crop recognition using PRISMA data: random forest (RF), artificial neural network (ANN), and convolutional neural network (CNN). The study was carried out over an area of 900 km2 in the province of Caserta, Italy. The PRISMA HDF5 file, pre-processed by the ASI at the reflectance level (L2d), was converted to GeoTiff using a custom Python script to facilitate its management in Qgis. The Qgis plugin AVHYAS was used for classification tests. The results show that CNN gives better results in terms of overall accuracy (0.973), K coefficient (0.968), and F1 score (0.842).

Hybrid system model for wind abrasion segmentation using semi-automatic classification of remote sensing multispectral areas

Wind abrasion, caused by particles transported by strong winds impacting on structures, can lead to their degradation. Although this phenomenon has hardly been studied in this context, it is becoming increasingly important due to new trends in infrastructure location, especially in renewable energy terms. Metallic structures are particularly vulnerable to degradation by the action of windblown sand particles. However, characterising such secluded sites is complicated, and remote sensing systems and satellite information become crucial. The objective of this research is to identify and delineate the geographic areas that are vulnerable to this phenomenon by employing a hybrid model with historical data and the semi-automatic classification of multispectral satellite images. The model is based on critical variables identified by the scientific community and case studies documented in the literature. The methodology used for the study consists of four phases, including creating a scientifically robust database, downloading and managing satellite and historical long-term information, segmenting the regions of interest, and modelling using supervised classification techniques. The proposed algorithm shows very accurate results (R2 = 0.9922) and the overall system approach is presented as a useful and generalizable method to address this problem, increasing the existing knowledge on material wear by particle action, and contributing to optimizing the initial design of resilient structures.

Arid Rangeland Degradation and Its Driving Forces in Southern Hodna, Algeria

Abstract The Algerian steppes frequently experience droughts and significant anthropogenic activities than harm its natural resources, endangering the fragile arid ecosystem’s balance. Hence, an analysis of steppe degradation can be instrumental in initiating remedial measures in the region. This study aims to assess the level of rangeland degradation in the southern Hodna region, a crucial component of the steppe landscape, and analyse the factors driving this degradation. The methodology employed a supervised classification technique, using multi-temporal remote sensing imagery from 1987 to 2016 to analyse land cover distribution. Furthermore, land cover dynamics were combined with carefully selected climatic and anthropogenic data, which were then subjected to canonical correspondece analysis to identify the primary drivers of rangeland degradation. The results reveal a sustained change in land cover dynamics, with a no table 48.65 % reduction in the rangeland class. Simultaneously, there has been a continuous expansion in sand dunes, rangeland degradation, and agriculture classes. It is worth noting that canonical correspondence analysis reveals stronger associations between rangeland degradation and anthropogenic factors than climatic factors. Finally, the study offers crucial insights into rangeland degradation in southern Hodna, aiding policymakers and land managers in developing sustainable development strategies for the region.

Modeling of land use/land cover dynamics using artificial neural network and cellular automata Markov chain algorithms in Goang watershed, Ethiopia

Land Use/Land Cover (LULC) change has inhibited sustainable development for the last millennia by affecting climate, biological cycles, and ecosystem services and functions. In this regard, understanding the historical and future patterns of LULC change plays a crucial role in implementing effective natural resource management. This study aimed to model and characterize the spatiotemporal trajectories of landscape change between the 1984 and 2060 periods. The satellite image spectral information was segmented into seven LULC classes using a hybrid approach of image spectral recognition. The supervised classification technique of Support Vector Machine (SVM) was used to classify the satellite images, whilst the Land Change Modeler (LCM) Module in TerrSet software was used to assess the historical trend and future simulation of LULC dynamics. To predict future landscape changes, transition potential maps were generated using a Multi-layer Perceptron (MLP) neural network algorithm. The findings of the study demonstrated that the Goang Watershed has experienced significant LULC change since 1984. During the 1984–2001, 2001–2022, and 1984–2022 periods, farmland showed a dramatic increasing trend with 7.5 km2/yr−1, 110.3 km2/yr−1, and 64.3 km2/yr−1, respectively. A similar trend was also observed in built-up areas with 0.5 km2/yr−1, 3.2 km2/yr−1, and 2 km2/yr−1. The expansion of farmland and built-up area was at the expense of forest, shrubland, and grasslands. With a business-as-usual scenario, the extent of farmland will continue to increase between 2022 and 2060 while rapid reduction is expected by forest, shrubland, and grasslands. The alarming rate of farmland and built-up area expansion will put significant pressure on biodiversity and ecosystem services in the area. As a result, eco-friendly conservation approaches should be implemented as soon as possible to maintain ecosystem health and encourage sustainable development.

ASSESSING THE EVOLUTION OF PADDY CULTIVATION IN KOTA BELUD, SABAH USING GIS AND REMOTE SENSING TECHNIQUES

This study aims to analyse the development of paddy cultivation in Kota Belud, Sabah using remote sensing and Geographic Information Systems (GIS) from 1990 to 2020. The cultivation of agriculture-based crops began in the early 1990s, and in 2017-2018, the Malaysian and state government identified Kota Belud as a paddy granary, making the cultivation of paddy a priority. This was in line with the National Agro-Food Policy 2021-2030, which aims to improve food security policies and reduce dependence on imports by enhancing rice production through technology reform and various initiatives. This study employed GIS and remote sensing techniques to analyse the changes in land use for paddy cultivation. Landsat TM 5 and Landsat TM 8 images were used to extract data of land use from 1990 to 2020. The results indicate that the area of paddy cultivation increased from 4,329 ha in 1990 to 12,564 ha in 2020, with fluctuations in between. The specific GIS and remote sensing techniques used in the analysis included unsupervised and supervised classification technique with accuracy classification of 94%, 86%, 98.30% and 91.60 % for year 1990, 2000, 2010 and 2010, respectively. Overall, the findings of this study can be used as a guideline by local authorities to improve rice production and food security in Malaysia.

Remote sensing of salmonid spawning sites in freshwater ecosystems: The potential of low-cost UAV data.

Salmonids are especially vulnerable during their embryonic development, but monitoring of their spawning grounds is rare and often relies on manual counting of their nests (redds). This method, however, is prone to sampling errors resulting in over- or underestimations of redd counts. Salmonid spawning habitat in shallow water areas can be distinguished by their visible reflection which makes the use of standard unmanned aerial vehicles (UAV) a viable option for their mapping. Here, we aimed to develop a standardised approach to detect salmonid spawning habitat that is easy and low-cost. We used a semi-automated approach by applying supervised classification techniques to UAV derived RGB imagery from two contrasting lakes in Iceland. For both lakes six endmember classes were obtained with high accuracies. Most importantly, producer's and user's accuracy for classifying spawning redds was >90% after applying post-classification improvements for both study areas. What we are proposing here is an entirely new approach for monitoring spawning habitats which will address some the major shortcomings of the widely used redd count method e.g. collecting and analysing large amounts of data cost and time efficiently, limiting observer bias, and allowing for precise quantification over different temporal and spatial scales.

The Land-Use and Land-Cover Changes in the Este District, South Gondar Zone, Northwestern Ethiopia, in the Last Four Decades (the 1980s to 2020s)

Environmental transformations are the results of land-use and land-cover (LULC) changes. This study aims to investigate the LULC changes and associated factors in the Este District in northwestern Ethiopia, for the last four decades (the 1980s to 2020s). The land-use and land-cover classes were analyzed using supervised classification techniques in ERDASS IMAGINE 2015 and ArcGIS 10.3.1, categorizing the Landsat satellite images for 1984, 2000, and 2018 into six cover classes: settlement, forests-shrublands, cropland, grassland, bare land, and water body. We used a historical Google map, topo-sheets, and ancillary data to verify the classification accuracies for 1984, 2000, and 2018, respectively. The climate and demographic data were obtained from the Ethiopian Meteorological Station and Ethiopian Statistical Agency, respectively. In addition, data from key informant interviews and focus group discussions were also used to understand the local community experiences and perceptions toward LULC changes. The values of each LULC class were correlated with the demographic and climatic variables, using simple correlation analysis to evaluate the impact of demographic and climatic conditions on LULC changes. The analysis indicated that the least and largest classes of LULC in Este District were water bodies (mean cover = 1.9 km2) and croplands (mean cover = 791.7 km2), respectively. Cropland coverage increased by 2% in 2000, then decreased by 11% in 2018. Between 1984 and 2018, the grasslands and settlements increased by 22% and 0.5%, respectively. Half of the bare land and one-tenth of the forests-shrublands also decreased over the last four decades (the 1980s to 2020s). As a result, the original croplands and forests–shrublands classes (11% each) before 2000 were converted into new grasslands and croplands after 2018, respectively. The study indicated that precipitation, solar radiation, and population growth are the potential drivers, and the perceptions of local communities are nearly in line with the statistical analysis results. Alternative income sources, such as tourism and carbon trading and the participation in afforestation programs, could reverse the situation in the study area.

Assessment of Land Use Land Cover Changes and Future Predictions Using CA-ANN Simulation for Gazipur City Corporation, Bangladesh

Anthropogenic activities have a significant influence on land use and land cover (LULC) changes, especially in rapidly growing areas. Among several models, the combination of a cellular automata–artificial neural network (CA-ANN) model is being widely used for assessing future LULC changes using satellite images. This study aimed to investigate LULC changes in Gazipur City Corporation (GCC), Bangladesh, and the changes in LULC patterns over the last two decades (2002 to 2022). In this study, the maximum likelihood supervised classification technique was used for processing the available satellite images. The results show that the urban area and vegetation coverage increased by 150% and 22.78%, whereas the bare land and waterbody decreased by 7.02% and 78.9%, respectively, from 2002 to 2022 inside the GCC area. For future LULC predictions, the CA-ANN model was developed, the accuracy percentage of which was 86.49%, and the kappa value was 0.83. The future LULC prediction model results show that the urban area will increase by 47.61%, whereas the bare land and waterbody are supposed to decrease by 24.17% and 67.23%, respectively, by 2042. The findings of this study could be useful for future sustainable urban planning and management, as well as enabling decision making by authorities for improvements in environmental and ecological conditions in the study area.

Sentiment analysis from Bangladeshi food delivery startup based on user reviews using machine learning and deep learning

Food delivery methods are at the top of the list in today's world. People's attitudes toward food delivery systems are usually influenced by food quality and delivery time. We did a sentiment analysis of consumer comments on the Facebook pages of Food Panda, HungryNaki, Pathao Food, and Shohoz Food, and data was acquired from these four sites’ remarks. In natural language processing (NLP) task, before the model was implemented, we went through a rigorous data pre-processing process that included stages like adding contractions, removing stop words, tokenizing, and more. Four supervised classification techniques are used: extreme gradient boosting (XGB), random forest classifier (RFC), decision tree classifier (DTC), and multi nominal Naive Bayes (MNB). Three deep learning (DL) models are used: convolutional neural network (CNN), long term short memory (LSTM), and recurrent neural network (RNN). The XGB model exceeds all four machine learning (ML) algorithms with an accuracy of 89.64%. LSTM has the highest accuracy rate of the three DL algorithms, with an accuracy of 91.07%. Among ML and DL models, LSTM DL takes the lead to predict the sentiment.

Sentiment analysis from Bangladeshi food delivery startup based on user reviews using machine learning and deep learning

Food delivery methods are at the top of the list in today's world. People's attitudes toward food delivery systems are usually influenced by food quality and delivery time. We did a sentiment analysis of consumer comments on the Facebook pages of Food Panda, HungryNaki, Pathao Food, and Shohoz Food, and data was acquired from these four sites’ remarks. In natural language processing (NLP) task, before the model was implemented, we went through a rigorous data pre-processing process that included stages like adding contractions, removing stop words, tokenizing, and more. Four supervised classification techniques are used: extreme gradient boosting (XGB), random forest classifier (RFC), decision tree classifier (DTC), and multi nominal Naive Bayes (MNB). Three deep learning (DL) models are used: convolutional neural network (CNN), long term short memory (LSTM), and recurrent neural network (RNN). The XGB model exceeds all four machine learning (ML) algorithms with an accuracy of 89.64%. LSTM has the highest accuracy rate of the three DL algorithms, with an accuracy of 91.07%. Among ML and DL models, LSTM DL takes the lead to predict the sentiment.

Mango Area Mapping Using Very High-Resolution Satellite Data in Major Blocks of Krishnagiri District, Tamil Nadu, India

A research study was carried out for mapping mango plantation from LISS IV data in the major blocks of Krishnagiri district, Tamil Nadu where there is substantial production of Mango. A very high-resolution satellite data namely LISS IV was acquired and processed with GIS tools. Ground truth data gathered during the survey were utilized to identify significant dB values for mango plantations, which were then used to classify the mango pixels in the study region using supervised classification technique. The Mango area in major blocks of Krishnagiri district was found to be 9077.9 ha during the year 2023. Accuracy assessment and cross validation was done using confusion matrix with the ground truth points collected. The classification resulted with an overall accuracy of 91.2 per cent with a kappa score of 0.62.