Mursal Furqan | NED University of Engineering and Technology (original) (raw)
Papers by Mursal Furqan
VOLUME-8 ISSUE-10, AUGUST 2019, REGULAR ISSUE
Heart disease is a common problem which can be very severe in old ages and also in people not hav... more Heart disease is a common problem which can be very severe in old ages and also in people not having a healthy lifestyle. With regular check-up and diagnosis in addition to maintaining a decent eating habit can prevent it to some extent. In this paper we have tried to implement the most sought after and important machine learning algorithm to predict the heart disease in a patient. The decision tree classifier is implemented based on the symptoms which are specifically the attributes required for the purpose of prediction. Using the decision tree algorithm, we will be able to identify those attributes which are the best one that will lead us to a better prediction of the datasets. The decision tree algorithm works in a way where it tries to solve the problem by the help of tree representation. Here each internal node of the tree represents an attribute, and each leaf node corresponds to a class label. The support vector machine algorithm helps us to classify the datasets on the basi...
2019 2nd International Conference on Computing, Mathematics and Engineering Technologies (iCoMET)
In this paper, an inexpensive, a lightweight and easily controlled robotic arm with Arduino contr... more In this paper, an inexpensive, a lightweight and easily controlled robotic arm with Arduino controlled structure is developed. The developed robotic arm is fully functional and accurately reacts to the movements from the user dictated via servo motors with respect to the required positions. The human user places the lightweight robotic claw at the desired place and uses the potentiometers to adjust the location of the claw, the arm and the belt. These potentiometers relay both movements, horizontal and vertical. The user can also control the robotic claw remotely via the network, built-in inside the Arduino board in the later versions. By using the networking Arduino board, one can use the claw from anywhere within the range and the user can also extend the wires to manually use the claw without any network support. The research methodology and design are introduced before the robot system is realized and tested. Finally, conclusions are given, and future improvements are discussed.
The world has seen an unprecedented and exponential increase in cases of heart disease worldwide ... more The world has seen an unprecedented and exponential increase in cases of heart disease worldwide every day. In the paper, the early prognosis of heart disease through careful treatment and the implementation of a healthy lifestyle through other studies will help prevent many cardiovascular diseases. This paper discusses a statistical model of heart disease that, based on basic parameters of the patients' health history, will help medical examiners and cardiac practitioners forecast heart disease. To build this prediction model, three (03) different Machine Learning Classifier Models are used, namely, Logistic Regression Classifier, K-Nearest Neighbours Classifier, and Random Forest Classifier. Different important clinical features of a patient, critical for deciding a patient's heart disease, are taken in the first section and, secondly, different ML Classifiers are defined on the given dataset and their accuracy calculated. For this analysis, the University College Irvine (...
INTERNATIONAL CONFERENCE ON COMPUTATIONAL SCIENCE AND TECH CONFERENCE, 2020
There have been many airline accidents especially in Pakistan, the nation has lost many precious ... more There have been many airline accidents especially in Pakistan, the nation has lost many precious lives and had a bad impact on the economy. This research study is about finding reasons and statistical analysis of airline accidents. There is not recent study available in this area from Pakistan Region, although the research studies are done on accident analysis of the international airline. The data science algorithms and the statistical analysis techniques are used to find out the reasons and data abnormalities. Mostly the model used in this study is based upon a supervised learning algorithm, i.e., Naïve based classifier (because it is very efficient and produces better predictions when there is fewer data in the dataset and in this case, dataset contains only 22 data entries) since it is used to classify analysis. Histogram analysis is used for graphical representation. The dataset used in this research is data of all airline accidents in Pakistan since 1947. The study gives the best results and helps to find out causes and suggests how accidents can be reduced.
INTERNATIONAL CONFERENCE ON COMPUTATIONAL SCIENCE AND TECH CONFERENCE 2020, 2020
The world has seen an unprecedented and exponential increase in cases of heart disease worldwide ... more The world has seen an unprecedented and exponential increase in cases of heart disease worldwide every day. In the paper, the early prognosis of heart disease through careful treatment and the implementation of a healthy lifestyle through other studies will help prevent many cardiovascular diseases. This paper discusses a statistical model of heart disease that, based on basic parameters of the patients' health history, will help medical examiners and cardiac practitioners forecast heart disease. To build this prediction model, three (03) different Machine Learning Classifier Models are used, namely, Logistic Regression Classifier, K-Nearest Neighbours Classifier, and Random Forest Classifier. Different important clinical features of a patient, critical for deciding a patient's heart disease, are taken in the first section and, secondly, different ML Classifiers are defined on the given dataset and their accuracy calculated. For this analysis, the University College Irvine (UCI) Dataset incorporates all the above attributes of possible heart attack patients.
2019 2nd International Conference on Computing, Mathematics and Engineering Technologies (iCoMET), 2019
In this paper, an inexpensive, a lightweight and easily controlled robotic arm with Arduino contr... more In this paper, an inexpensive, a lightweight and easily controlled robotic arm with Arduino controlled structure is developed. The developed robotic arm is fully functional and accurately reacts to the movements from the user dictated via servo motors with respect to the required positions. The human user places the lightweight robotic claw at the desired place and uses the potentiometers to adjust the location of the claw, the arm and the belt. These potentiometers relay both movements, horizontal and vertical. The user can also control the robotic claw remotely via the network, built-in inside the Arduino board in the later versions. By using the networking Arduino board, one can use the claw from anywhere within the range and the user can also extend the wires to manually use the claw without any network support. The research methodology and design are introduced before the robot system is realized and tested. Finally, conclusions are given, and future improvements are discussed.
VOLUME-8 ISSUE-10, AUGUST 2019, REGULAR ISSUE
Heart disease is a common problem which can be very severe in old ages and also in people not hav... more Heart disease is a common problem which can be very severe in old ages and also in people not having a healthy lifestyle. With regular check-up and diagnosis in addition to maintaining a decent eating habit can prevent it to some extent. In this paper we have tried to implement the most sought after and important machine learning algorithm to predict the heart disease in a patient. The decision tree classifier is implemented based on the symptoms which are specifically the attributes required for the purpose of prediction. Using the decision tree algorithm, we will be able to identify those attributes which are the best one that will lead us to a better prediction of the datasets. The decision tree algorithm works in a way where it tries to solve the problem by the help of tree representation. Here each internal node of the tree represents an attribute, and each leaf node corresponds to a class label. The support vector machine algorithm helps us to classify the datasets on the basi...
2019 2nd International Conference on Computing, Mathematics and Engineering Technologies (iCoMET)
In this paper, an inexpensive, a lightweight and easily controlled robotic arm with Arduino contr... more In this paper, an inexpensive, a lightweight and easily controlled robotic arm with Arduino controlled structure is developed. The developed robotic arm is fully functional and accurately reacts to the movements from the user dictated via servo motors with respect to the required positions. The human user places the lightweight robotic claw at the desired place and uses the potentiometers to adjust the location of the claw, the arm and the belt. These potentiometers relay both movements, horizontal and vertical. The user can also control the robotic claw remotely via the network, built-in inside the Arduino board in the later versions. By using the networking Arduino board, one can use the claw from anywhere within the range and the user can also extend the wires to manually use the claw without any network support. The research methodology and design are introduced before the robot system is realized and tested. Finally, conclusions are given, and future improvements are discussed.
The world has seen an unprecedented and exponential increase in cases of heart disease worldwide ... more The world has seen an unprecedented and exponential increase in cases of heart disease worldwide every day. In the paper, the early prognosis of heart disease through careful treatment and the implementation of a healthy lifestyle through other studies will help prevent many cardiovascular diseases. This paper discusses a statistical model of heart disease that, based on basic parameters of the patients' health history, will help medical examiners and cardiac practitioners forecast heart disease. To build this prediction model, three (03) different Machine Learning Classifier Models are used, namely, Logistic Regression Classifier, K-Nearest Neighbours Classifier, and Random Forest Classifier. Different important clinical features of a patient, critical for deciding a patient's heart disease, are taken in the first section and, secondly, different ML Classifiers are defined on the given dataset and their accuracy calculated. For this analysis, the University College Irvine (...
INTERNATIONAL CONFERENCE ON COMPUTATIONAL SCIENCE AND TECH CONFERENCE, 2020
There have been many airline accidents especially in Pakistan, the nation has lost many precious ... more There have been many airline accidents especially in Pakistan, the nation has lost many precious lives and had a bad impact on the economy. This research study is about finding reasons and statistical analysis of airline accidents. There is not recent study available in this area from Pakistan Region, although the research studies are done on accident analysis of the international airline. The data science algorithms and the statistical analysis techniques are used to find out the reasons and data abnormalities. Mostly the model used in this study is based upon a supervised learning algorithm, i.e., Naïve based classifier (because it is very efficient and produces better predictions when there is fewer data in the dataset and in this case, dataset contains only 22 data entries) since it is used to classify analysis. Histogram analysis is used for graphical representation. The dataset used in this research is data of all airline accidents in Pakistan since 1947. The study gives the best results and helps to find out causes and suggests how accidents can be reduced.
INTERNATIONAL CONFERENCE ON COMPUTATIONAL SCIENCE AND TECH CONFERENCE 2020, 2020
The world has seen an unprecedented and exponential increase in cases of heart disease worldwide ... more The world has seen an unprecedented and exponential increase in cases of heart disease worldwide every day. In the paper, the early prognosis of heart disease through careful treatment and the implementation of a healthy lifestyle through other studies will help prevent many cardiovascular diseases. This paper discusses a statistical model of heart disease that, based on basic parameters of the patients' health history, will help medical examiners and cardiac practitioners forecast heart disease. To build this prediction model, three (03) different Machine Learning Classifier Models are used, namely, Logistic Regression Classifier, K-Nearest Neighbours Classifier, and Random Forest Classifier. Different important clinical features of a patient, critical for deciding a patient's heart disease, are taken in the first section and, secondly, different ML Classifiers are defined on the given dataset and their accuracy calculated. For this analysis, the University College Irvine (UCI) Dataset incorporates all the above attributes of possible heart attack patients.
2019 2nd International Conference on Computing, Mathematics and Engineering Technologies (iCoMET), 2019
In this paper, an inexpensive, a lightweight and easily controlled robotic arm with Arduino contr... more In this paper, an inexpensive, a lightweight and easily controlled robotic arm with Arduino controlled structure is developed. The developed robotic arm is fully functional and accurately reacts to the movements from the user dictated via servo motors with respect to the required positions. The human user places the lightweight robotic claw at the desired place and uses the potentiometers to adjust the location of the claw, the arm and the belt. These potentiometers relay both movements, horizontal and vertical. The user can also control the robotic claw remotely via the network, built-in inside the Arduino board in the later versions. By using the networking Arduino board, one can use the claw from anywhere within the range and the user can also extend the wires to manually use the claw without any network support. The research methodology and design are introduced before the robot system is realized and tested. Finally, conclusions are given, and future improvements are discussed.