Abstract— This paper proffers a secured and cost effective<br> model for user behavioural a... more Abstract— This paper proffers a secured and cost effective<br> model for user behavioural analysis (UBA) in smart homes. The<br> model was built with support vector machine algorithm on a<br> generated dataset on constructed prototype of smart home. The<br> prototype had motion sensors and actuators for data collection and<br> switching on and off of things in the home. Another model called<br> human activity recognition (HAR) was developed with Gaussian<br> naïve bayes to interface with UBA for collection of information<br> from the home. The processor of the system is raspberry pi 4<br> computer board that runs on Linux-based Raspbian Operating<br> System. The two models earlier mentioned were integrated in the<br> processor. The information generated from the system is processed<br> in the processor and disseminated to things in the home. The Blynk<br> Mobile App was used to interface with the integrate...
Patients waste great deal of resources in the cause of identification of pathogens that caused th... more Patients waste great deal of resources in the cause of identification of pathogens that caused their ailments; this calls for concern, hence the need to develop a veritable tool for minimizing the cost involved in classification of disease pathogens without compromising accuracy. In this paper, we developed a feature extraction model which reduces the clinical markers for prostate cancer and diabetes. The feature extraction, in the form of principal component analysis (PCA), was used to extract relevant components from prostate cancer and diabetes datasets. The simulation and experiment of the system were done with matlab. The system was able to extract 3 relevant features out of 4 prostate cancer clinical markers and 4 relevant features out of 5 diabetes clinical markers. The result showed that when trained in a multilayer neural network it yielded better classification accuracy with the extracted relevant features with 80% and 75% component analysis in prostate cancer and diabetes...
The advancement in Information Technology (IT) has assisted in inculcating the three Nigeria majo... more The advancement in Information Technology (IT) has assisted in inculcating the three Nigeria major languages in text-based application such as text mining, information retrieval and natural language processing. The interest of this paper is the Igbo language, which uses compounding as a common type of word formation and as well has many vocabularies of compound words. The issues of collocation, word ordering and compounding play high role in Igbo language. The ambiguity in dealing with these compound words has made the representation of Igbo language text document very difficult because this cannot be addressed using the most common and standard approach of the Bag-Of-Words (BOW) model of text representation, which ignores the word order and relation. However, this cause for a concern and the need to develop an improved model to capture this situation. This paper presents the analysis of Igbo language text document, considering its compounding nature and describes its representation...
International Journal of Data Mining Techniques and Applications
The advancement in Information Technology (IT) has assisted in inculcating the three Nigeria majo... more The advancement in Information Technology (IT) has assisted in inculcating the three Nigeria major languages in text-based application such as text mining, information retrieval and natural language processing. The interest of this paper is the Igbo language, which uses compounding as a common type of word formation and as well has many vocabularies of compound words. The issues of collocation, word ordering and compounding play high role in Igbo language. The ambiguity in dealing with these compound words has made the representation of Igbo language text document very difficult because this cannot be addressed using the most common and standard approach of the Bag-Of-Words (BOW) model of text representation, which ignores the word order and relation. However, this cause for a concern and the need to develop an improved model to capture this situation. This paper presents the analysis of Igbo language text document, considering its compounding nature and describes its representation...
Misguided information in health care has caused much havoc that have led to the death of millions... more Misguided information in health care has caused much havoc that have led to the death of millions of people as a result of misclassification, and inconsistent health care records; hence the objective of this paper is to develop an improved clinical decision support system. This system incorporated hybrid system of non-knowledge based and knowledge based decision support system for the diagnosis of diseases and proper health care delivery records using prostate cancer and diabetes datasets to train and validate the model. The min-max method was adopted in normalizing the datasets, while genetic algorithm was deployed in initiating the training weights of the MLP. The result obtained in this paper yielded a classification accuracy of 98%, sensitivity of 0.98 and specificity of 100 for prostate cancer and accuracy of 94%, sensitivity of 0.94 and specificity of 0.67 for diabetes
The automatic classification of legal case documents has become very important, owing to the just... more The automatic classification of legal case documents has become very important, owing to the justice denials, delays and failures observed in the judicial case management systems. Our hybrid text classification model employed extensive preprocessing techniques to prepare the document features, the probabilistic nature of the Naïve Bayes algorithm was integrated to generate vectorized data from the document features for the classifier, and the most important features was selected by feature ranking using the Chi Square method for final classification using the Support Vector Machine. The hybrid text classifier application was implemented using the Object Oriented Analysis and Design Methodology and developed using the Java programming language and MySQL. Results showed that best features were selected and the documents were accurately classified to their right categories using this hybrid application, as proven using standard performance measure metrics.
The smart-phone industry has witnessed tremendous growth in recent history simply because of the ... more The smart-phone industry has witnessed tremendous growth in recent history simply because of the emergence of the android operating system. It is now easier to make payments on our mobile phones but one major hindrance lies in transaction security. The objective of this paper is to develop a secure transaction protocol for an android based mobile payment system using quick response code technology and a hybrid cryptographic scheme. To achieve a better security in the system, we implemented symmetric, asymmetric cryptography alongside hashing and message authentication codes simultaneously in the system protocol. The results obtained depict a secure mobile payment system which makes use of dual authentication mechanism by two distinct entities
The emergence of Android mobile operating system has brought about a huge change in the smart pho... more The emergence of Android mobile operating system has brought about a huge change in the smart phone industry. It has contributed
to the increase in the range of services a smart phone can provide which includes the ability to be used in the exchange of financial
value. Quick Response (QR) code technology can be regarded as a key technology in the future of mobile payments. The goal of this
paper is to develop a model of mobile payment solution using QR code as a communication link between the two parties engaging
in a transaction and also for encryption of transaction details. The study resulted in a new android mobile wallet application which
was developed using the object oriented analysis and design methodology. Our experimental results show that the new system has
better security using dual authentication employed by different entities, does not require sharing of personal details, eliminates the
need for agents in the payment ecosystem and satisfies the universal property
Abstract— This paper proffers a secured and cost effective<br> model for user behavioural a... more Abstract— This paper proffers a secured and cost effective<br> model for user behavioural analysis (UBA) in smart homes. The<br> model was built with support vector machine algorithm on a<br> generated dataset on constructed prototype of smart home. The<br> prototype had motion sensors and actuators for data collection and<br> switching on and off of things in the home. Another model called<br> human activity recognition (HAR) was developed with Gaussian<br> naïve bayes to interface with UBA for collection of information<br> from the home. The processor of the system is raspberry pi 4<br> computer board that runs on Linux-based Raspbian Operating<br> System. The two models earlier mentioned were integrated in the<br> processor. The information generated from the system is processed<br> in the processor and disseminated to things in the home. The Blynk<br> Mobile App was used to interface with the integrate...
Patients waste great deal of resources in the cause of identification of pathogens that caused th... more Patients waste great deal of resources in the cause of identification of pathogens that caused their ailments; this calls for concern, hence the need to develop a veritable tool for minimizing the cost involved in classification of disease pathogens without compromising accuracy. In this paper, we developed a feature extraction model which reduces the clinical markers for prostate cancer and diabetes. The feature extraction, in the form of principal component analysis (PCA), was used to extract relevant components from prostate cancer and diabetes datasets. The simulation and experiment of the system were done with matlab. The system was able to extract 3 relevant features out of 4 prostate cancer clinical markers and 4 relevant features out of 5 diabetes clinical markers. The result showed that when trained in a multilayer neural network it yielded better classification accuracy with the extracted relevant features with 80% and 75% component analysis in prostate cancer and diabetes...
The advancement in Information Technology (IT) has assisted in inculcating the three Nigeria majo... more The advancement in Information Technology (IT) has assisted in inculcating the three Nigeria major languages in text-based application such as text mining, information retrieval and natural language processing. The interest of this paper is the Igbo language, which uses compounding as a common type of word formation and as well has many vocabularies of compound words. The issues of collocation, word ordering and compounding play high role in Igbo language. The ambiguity in dealing with these compound words has made the representation of Igbo language text document very difficult because this cannot be addressed using the most common and standard approach of the Bag-Of-Words (BOW) model of text representation, which ignores the word order and relation. However, this cause for a concern and the need to develop an improved model to capture this situation. This paper presents the analysis of Igbo language text document, considering its compounding nature and describes its representation...
International Journal of Data Mining Techniques and Applications
The advancement in Information Technology (IT) has assisted in inculcating the three Nigeria majo... more The advancement in Information Technology (IT) has assisted in inculcating the three Nigeria major languages in text-based application such as text mining, information retrieval and natural language processing. The interest of this paper is the Igbo language, which uses compounding as a common type of word formation and as well has many vocabularies of compound words. The issues of collocation, word ordering and compounding play high role in Igbo language. The ambiguity in dealing with these compound words has made the representation of Igbo language text document very difficult because this cannot be addressed using the most common and standard approach of the Bag-Of-Words (BOW) model of text representation, which ignores the word order and relation. However, this cause for a concern and the need to develop an improved model to capture this situation. This paper presents the analysis of Igbo language text document, considering its compounding nature and describes its representation...
Misguided information in health care has caused much havoc that have led to the death of millions... more Misguided information in health care has caused much havoc that have led to the death of millions of people as a result of misclassification, and inconsistent health care records; hence the objective of this paper is to develop an improved clinical decision support system. This system incorporated hybrid system of non-knowledge based and knowledge based decision support system for the diagnosis of diseases and proper health care delivery records using prostate cancer and diabetes datasets to train and validate the model. The min-max method was adopted in normalizing the datasets, while genetic algorithm was deployed in initiating the training weights of the MLP. The result obtained in this paper yielded a classification accuracy of 98%, sensitivity of 0.98 and specificity of 100 for prostate cancer and accuracy of 94%, sensitivity of 0.94 and specificity of 0.67 for diabetes
The automatic classification of legal case documents has become very important, owing to the just... more The automatic classification of legal case documents has become very important, owing to the justice denials, delays and failures observed in the judicial case management systems. Our hybrid text classification model employed extensive preprocessing techniques to prepare the document features, the probabilistic nature of the Naïve Bayes algorithm was integrated to generate vectorized data from the document features for the classifier, and the most important features was selected by feature ranking using the Chi Square method for final classification using the Support Vector Machine. The hybrid text classifier application was implemented using the Object Oriented Analysis and Design Methodology and developed using the Java programming language and MySQL. Results showed that best features were selected and the documents were accurately classified to their right categories using this hybrid application, as proven using standard performance measure metrics.
The smart-phone industry has witnessed tremendous growth in recent history simply because of the ... more The smart-phone industry has witnessed tremendous growth in recent history simply because of the emergence of the android operating system. It is now easier to make payments on our mobile phones but one major hindrance lies in transaction security. The objective of this paper is to develop a secure transaction protocol for an android based mobile payment system using quick response code technology and a hybrid cryptographic scheme. To achieve a better security in the system, we implemented symmetric, asymmetric cryptography alongside hashing and message authentication codes simultaneously in the system protocol. The results obtained depict a secure mobile payment system which makes use of dual authentication mechanism by two distinct entities
The emergence of Android mobile operating system has brought about a huge change in the smart pho... more The emergence of Android mobile operating system has brought about a huge change in the smart phone industry. It has contributed
to the increase in the range of services a smart phone can provide which includes the ability to be used in the exchange of financial
value. Quick Response (QR) code technology can be regarded as a key technology in the future of mobile payments. The goal of this
paper is to develop a model of mobile payment solution using QR code as a communication link between the two parties engaging
in a transaction and also for encryption of transaction details. The study resulted in a new android mobile wallet application which
was developed using the object oriented analysis and design methodology. Our experimental results show that the new system has
better security using dual authentication employed by different entities, does not require sharing of personal details, eliminates the
need for agents in the payment ecosystem and satisfies the universal property
Uploads
Papers by Dr Ugwu Chidiebere
and failures observed in the judicial case management systems.
Our hybrid text classification model employed extensive preprocessing techniques to prepare the document features, the probabilistic nature of the Naïve Bayes algorithm was integrated to generate vectorized data from the document features for the classifier, and the most important features was selected by feature ranking using the Chi Square method for final classification using the Support Vector Machine. The hybrid text classifier application was implemented using the Object Oriented Analysis and Design Methodology and developed using the Java programming language and MySQL. Results showed that best features were selected and the documents were accurately classified to their right categories using this hybrid application, as proven using standard performance measure metrics.
to the increase in the range of services a smart phone can provide which includes the ability to be used in the exchange of financial
value. Quick Response (QR) code technology can be regarded as a key technology in the future of mobile payments. The goal of this
paper is to develop a model of mobile payment solution using QR code as a communication link between the two parties engaging
in a transaction and also for encryption of transaction details. The study resulted in a new android mobile wallet application which
was developed using the object oriented analysis and design methodology. Our experimental results show that the new system has
better security using dual authentication employed by different entities, does not require sharing of personal details, eliminates the
need for agents in the payment ecosystem and satisfies the universal property
and failures observed in the judicial case management systems.
Our hybrid text classification model employed extensive preprocessing techniques to prepare the document features, the probabilistic nature of the Naïve Bayes algorithm was integrated to generate vectorized data from the document features for the classifier, and the most important features was selected by feature ranking using the Chi Square method for final classification using the Support Vector Machine. The hybrid text classifier application was implemented using the Object Oriented Analysis and Design Methodology and developed using the Java programming language and MySQL. Results showed that best features were selected and the documents were accurately classified to their right categories using this hybrid application, as proven using standard performance measure metrics.
to the increase in the range of services a smart phone can provide which includes the ability to be used in the exchange of financial
value. Quick Response (QR) code technology can be regarded as a key technology in the future of mobile payments. The goal of this
paper is to develop a model of mobile payment solution using QR code as a communication link between the two parties engaging
in a transaction and also for encryption of transaction details. The study resulted in a new android mobile wallet application which
was developed using the object oriented analysis and design methodology. Our experimental results show that the new system has
better security using dual authentication employed by different entities, does not require sharing of personal details, eliminates the
need for agents in the payment ecosystem and satisfies the universal property