Abubakar Isa - Academia.edu (original) (raw)

Papers by Abubakar Isa

Research paper thumbnail of Digital Twins Temporal Dependencies-Based on Time Series Using Multivariate Long Short-Term Memory

Electronics

Digital Twins, which are virtual representations of physical systems mirroring their behavior, en... more Digital Twins, which are virtual representations of physical systems mirroring their behavior, enable real-time monitoring, analysis, and optimization. Understanding and identifying the temporal dependencies included in the multivariate time series data that characterize the behavior of the system are crucial for improving the effectiveness of Digital Twins. Long Short-Term Memory (LSTM) networks have been used to represent complex temporal dependencies and identify long-term links in the Industrial Internet of Things (IIoT). This paper proposed a Digital Twin temporal dependency technique using LSTM to capture the long-term dependencies in IIoT time series data, estimate the lag between the input and intended output, and handle missing data. Autocorrelation analysis showed the lagged links between variables, aiding in the discovery of temporal dependencies. The system evaluated the LSTM model by providing it with a set of previous observations and asking it to forecast the value at...

Research paper thumbnail of Design and implementation of microcontroller based PWM technique for sine wave inverter

2009 International Conference on Power Engineering, Energy and Electrical Drives, 2009

Abstract— A microcontroller based advanced technique of generating sine wave with minimized harmo... more Abstract— A microcontroller based advanced technique of generating sine wave with minimized harmonics is implemented in this paper. The proposed technique “impulse-sine product” aims to design and implement a voltage regulated inverter with ripple free and glitch free ...

Research paper thumbnail of Re-emergence of Indigeneity in transformed Layouts in Urban Public Housing in Nigeria

International Journal of Built Environment and Sustainability, 2016

In contrast to the one whole homogenous configuration provided by developers that has often prove... more In contrast to the one whole homogenous configuration provided by developers that has often proved ineffective overtime, transformed public housing units are widespread in developing cities. Though transformation is a natural phenomenon of inhabitation, its various levels and degree can have implicit values. In recent times in Nigeria, heterogeneous spatial patterns of transformed houses in public housing estates has given empirical evidence of essentially inevitable reflection of culture sensitivity as hypothesized by previous studies. That hints that the mainstream sustainable values were probably ignored during design process. This study attempted to connect that possible missing link. To achieve that aim, it developed two objectives. First, is to identify core values behind space layouts in the mainstream. Then, to examine how they exist in the transformed layout as reflected at communal origins. It investigated on 42 transformed units in selected urban Public housings neighborh...

Research paper thumbnail of Improved ANN-based algorithm for detection and classification of faults on transmission lines

2016 1st India International Conference on Information Processing (IICIP), 2016

This paper presents a relaying algorithm based on Artificial Neural Network (ANN) technique for t... more This paper presents a relaying algorithm based on Artificial Neural Network (ANN) technique for the protection of transmission line. A feed forward ANN with six inputs and eleven outputs has been developed for the detection and classification of faults. Data was generated by simulating a 400 kV, 50Hz, 100 km transmission line in PSCAD/EMTDC at a sampling frequency of 2 kHz. Three ANN configurations with different combinations of inputs have been attempted. Initially all the three ANN configurations were trained and tested using truncated data for their comparative performance. ANN-2 configuration has been found to be the best one and has an accuracy of 100% for fault detection and classification in both training and testing phases with the relay operating time of 12.5 ms. ANN-2 has been further trained and tested using full data. Two-fold cross verification was carried out. An accuracy of 100% was obtained on testing with 12.5 ms delay each time.

Research paper thumbnail of A More Comprehensive Method for Using The Target-side Monolingual Data to Improve Low Resource Neural Machine Translation

arXiv: Computation and Language, 2020

Improving neural machine translation (NMT) models using the back-translations of the monolingual ... more Improving neural machine translation (NMT) models using the back-translations of the monolingual target data (synthetic parallel data) is currently the state-of-the-art approach for training improved translation systems. The target-side side monolingual data has been used in the back-translation approach to improve the forward (target) translation model. Whereas the success of the approach heavily relies on the additional parallel data generating model -- the backward model -- the aim of the approach is only targeted at improving the forward model. The back-translation approach was designed primarily to benefit from an additional data whose source-side is synthetic. But research works have shown that translation models can also benefit from additional data whose target-side is synthetic. This work proposes the use of the target-side data throughout the back-translation approach to improve both the backward and forward models. We explored using only the target-side monolingual data t...

Research paper thumbnail of Interventions to prevent emergency department visit related to complementary and alternative medicines use

Journal of Complementary Medicine Research, 2018

Research paper thumbnail of Enhanced back-translation for low resource neural machine translation using self-training

Improving neural machine translation (NMT) models using the back-translations of the monolingual ... more Improving neural machine translation (NMT) models using the back-translations of the monolingual target data (synthetic parallel data) is currently the state-of-the-art approach for training improved translation systems. The quality of the backward system - which is trained on the available parallel data and used for the back-translation - has been shown in many studies to affect the performance of the final NMT model. In low resource conditions, the available parallel data is usually not enough to train a backward model that can produce the qualitative synthetic data needed to train a standard translation model. This work proposes a self-training strategy where the output of the backward model is used to improve the model itself through the forward translation technique. The technique was shown to improve baseline low resource IWSLT'14 English-German and IWSLT'15 English-Vietnamese backward translation models by 11.06 and 1.5 BLEUs respectively. The synthetic data generated...

Research paper thumbnail of Iterative Batch Back-Translation for Neural Machine Translation: A Conceptual Model

ArXiv, 2020

An effective method to generate a large number of parallel sentences for training improved neural... more An effective method to generate a large number of parallel sentences for training improved neural machine translation (NMT) systems is the use of back-translations of the target-side monolingual data. Recently, iterative back-translation has been shown to outperform standard back-translation albeit on some language pairs. This work proposes the iterative batch back-translation that is aimed at enhancing the standard iterative back-translation and enabling the efficient utilization of more monolingual data. After each iteration, improved back-translations of new sentences are added to the parallel data that will be used to train the final forward model. The work presents a conceptual model of the proposed approach.

Research paper thumbnail of A Hybrid Approach for Improved Low Resource Neural Machine Translation using Monolingual Data

arXiv: Computation and Language, 2020

Many language pairs are low resource, meaning the amount and/or quality of available parallel dat... more Many language pairs are low resource, meaning the amount and/or quality of available parallel data is not sufficient to train a neural machine translation (NMT) model which can reach an acceptable standard of accuracy. Many works have explored using the readily available monolingual data in either or both of the languages to improve the standard of translation models in low, and even high, resource languages. One of the most successful of such works is the back-translation that utilizes the translations of the target language monolingual data to increase the amount of the training data. The quality of the backward model which is trained on the available parallel data has been shown to determine the performance of the back-translation approach. Despite this, only the forward model is improved on the monolingual target data in standard back-translation. A previous study proposed an iterative back-translation approach for improving both models over several iterations. But unlike in the...

Research paper thumbnail of Design and Comparison Migration between Ipv4 and Ipv6 Transition Techniques

IPv4 which is the old version of Internet Protocol has a new successor named IP Next Generation (... more IPv4 which is the old version of Internet Protocol has a new successor named IP Next Generation (IPng) or IPv6 developed by Internet Engineering Task Force (IETF). This new version is developed specifically to resolve some issues of IPv4 like scalability, performance and reliability. Although new version is ready for usage, it is obvious that it will take years to transit fully from IPv4 to IPv6. We have to use these two versions together for a long time. Therefore, we have to investigate and know transition mechanisms that we can use during transition period to achieve a transition with minimum problem. This research defines the essential information about compatibility transition mechanisms between IPv4-IPv6. Dual Stack is one of the IPv4-IPv6 compatible mechanism by running both IPv4 stack and IPv6 stack in a single node. 6 to 4 tunneling mechanism encrypts IPv6 packets in IPv4 packets to make communications possible, from IPv6 network over IPv4 network. This has been configured ...

Research paper thumbnail of Enhanced Back-Translation for Low Resource Neural Machine Translation Using Self-training

Communications in Computer and Information Science

Improving neural machine translation (NMT) models using the back-translations of the monolingual ... more Improving neural machine translation (NMT) models using the back-translations of the monolingual target data (synthetic parallel data) is currently the state-of-the-art approach for training improved translation systems. The quality of the backward system-which is trained on the available parallel data and used for the back-translation-has been shown in many studies to affect the performance of the final NMT model. In low resource conditions, the available parallel data is usually not enough to train a backward model that can produce the qualitative synthetic data needed to train a standard translation model. This work proposes a self-training strategy where the output of the backward model is used to improve the model itself through the forward translation technique. The technique was shown to improve baseline low resource IWSLT'14 English-German and IWSLT'15 English-Vietnamese backward translation models by 11.06 and 1.5 BLEUs respectively. The synthetic data generated by the improved English-German backward model was used to train a forward model which out-performed another forward model trained using standard back-translation by 2.7 BLEU.

Research paper thumbnail of Using Self-Training to Improve Back-Translation in Low Resource Neural Machine Translation

ArXiv, 2020

Improving neural machine translation (NMT) models using the back-translations of the monolingual ... more Improving neural machine translation (NMT) models using the back-translations of the monolingual target data (synthetic parallel data) is currently the state-of-the-art approach for training improved translation systems. The quality of the backward system - which is trained on the available parallel data and used for the back-translation - has been shown in many studies to affect the performance of the final NMT model. In low resource conditions, the available parallel data is usually not enough to train a backward model that can produce the qualitative synthetic data needed to train a standard translation model. This work proposes a self-training strategy where the output of the backward model is used to improve the model itself through the forward translation technique. The technique was shown to improve baseline low resource IWSLT'14 English-German and IWSLT'15 English-Vietnamese backward translation models by 11.06 and 1.5 BLEUs respectively. The synthetic data generated...

Research paper thumbnail of Implementation of Microcontroller Based Distance Relay

European Journal of Electrical Engineering and Computer Science

In this project microcontroller 8051 was used for comparisons between set Impedance and calculate... more In this project microcontroller 8051 was used for comparisons between set Impedance and calculated Impedance for based distance relay of the protection of transmission line, based on that trip signal is send to circuit breaker. The Hardware consisted of main CT and VT, auxiliary VT and CT ,BLF ,sample and hold(S/H) circuit ,multiplexer ,ADC and microcontroller 8051.In which the main CT and VT step down the current and voltage respectably passed to auxiliary CT and VT for protecting the remaining components and convert the current into voltage in the output of auxiliary CT, then passed it to band limit filter to limit the frequency for sampling hold after sampling and hold circuit the signal is passed to multiplexer then to the ADC for analogue to digital conversion, after the conversion the digital signals is passed to the 8051 microcontroller to processing the signals that is calculating the average, impedance and compare it with the set impedance (store impedance) to generate tr...

Research paper thumbnail of Digital Twins Temporal Dependencies-Based on Time Series Using Multivariate Long Short-Term Memory

Electronics

Digital Twins, which are virtual representations of physical systems mirroring their behavior, en... more Digital Twins, which are virtual representations of physical systems mirroring their behavior, enable real-time monitoring, analysis, and optimization. Understanding and identifying the temporal dependencies included in the multivariate time series data that characterize the behavior of the system are crucial for improving the effectiveness of Digital Twins. Long Short-Term Memory (LSTM) networks have been used to represent complex temporal dependencies and identify long-term links in the Industrial Internet of Things (IIoT). This paper proposed a Digital Twin temporal dependency technique using LSTM to capture the long-term dependencies in IIoT time series data, estimate the lag between the input and intended output, and handle missing data. Autocorrelation analysis showed the lagged links between variables, aiding in the discovery of temporal dependencies. The system evaluated the LSTM model by providing it with a set of previous observations and asking it to forecast the value at...

Research paper thumbnail of Design and implementation of microcontroller based PWM technique for sine wave inverter

2009 International Conference on Power Engineering, Energy and Electrical Drives, 2009

Abstract— A microcontroller based advanced technique of generating sine wave with minimized harmo... more Abstract— A microcontroller based advanced technique of generating sine wave with minimized harmonics is implemented in this paper. The proposed technique “impulse-sine product” aims to design and implement a voltage regulated inverter with ripple free and glitch free ...

Research paper thumbnail of Re-emergence of Indigeneity in transformed Layouts in Urban Public Housing in Nigeria

International Journal of Built Environment and Sustainability, 2016

In contrast to the one whole homogenous configuration provided by developers that has often prove... more In contrast to the one whole homogenous configuration provided by developers that has often proved ineffective overtime, transformed public housing units are widespread in developing cities. Though transformation is a natural phenomenon of inhabitation, its various levels and degree can have implicit values. In recent times in Nigeria, heterogeneous spatial patterns of transformed houses in public housing estates has given empirical evidence of essentially inevitable reflection of culture sensitivity as hypothesized by previous studies. That hints that the mainstream sustainable values were probably ignored during design process. This study attempted to connect that possible missing link. To achieve that aim, it developed two objectives. First, is to identify core values behind space layouts in the mainstream. Then, to examine how they exist in the transformed layout as reflected at communal origins. It investigated on 42 transformed units in selected urban Public housings neighborh...

Research paper thumbnail of Improved ANN-based algorithm for detection and classification of faults on transmission lines

2016 1st India International Conference on Information Processing (IICIP), 2016

This paper presents a relaying algorithm based on Artificial Neural Network (ANN) technique for t... more This paper presents a relaying algorithm based on Artificial Neural Network (ANN) technique for the protection of transmission line. A feed forward ANN with six inputs and eleven outputs has been developed for the detection and classification of faults. Data was generated by simulating a 400 kV, 50Hz, 100 km transmission line in PSCAD/EMTDC at a sampling frequency of 2 kHz. Three ANN configurations with different combinations of inputs have been attempted. Initially all the three ANN configurations were trained and tested using truncated data for their comparative performance. ANN-2 configuration has been found to be the best one and has an accuracy of 100% for fault detection and classification in both training and testing phases with the relay operating time of 12.5 ms. ANN-2 has been further trained and tested using full data. Two-fold cross verification was carried out. An accuracy of 100% was obtained on testing with 12.5 ms delay each time.

Research paper thumbnail of A More Comprehensive Method for Using The Target-side Monolingual Data to Improve Low Resource Neural Machine Translation

arXiv: Computation and Language, 2020

Improving neural machine translation (NMT) models using the back-translations of the monolingual ... more Improving neural machine translation (NMT) models using the back-translations of the monolingual target data (synthetic parallel data) is currently the state-of-the-art approach for training improved translation systems. The target-side side monolingual data has been used in the back-translation approach to improve the forward (target) translation model. Whereas the success of the approach heavily relies on the additional parallel data generating model -- the backward model -- the aim of the approach is only targeted at improving the forward model. The back-translation approach was designed primarily to benefit from an additional data whose source-side is synthetic. But research works have shown that translation models can also benefit from additional data whose target-side is synthetic. This work proposes the use of the target-side data throughout the back-translation approach to improve both the backward and forward models. We explored using only the target-side monolingual data t...

Research paper thumbnail of Interventions to prevent emergency department visit related to complementary and alternative medicines use

Journal of Complementary Medicine Research, 2018

Research paper thumbnail of Enhanced back-translation for low resource neural machine translation using self-training

Improving neural machine translation (NMT) models using the back-translations of the monolingual ... more Improving neural machine translation (NMT) models using the back-translations of the monolingual target data (synthetic parallel data) is currently the state-of-the-art approach for training improved translation systems. The quality of the backward system - which is trained on the available parallel data and used for the back-translation - has been shown in many studies to affect the performance of the final NMT model. In low resource conditions, the available parallel data is usually not enough to train a backward model that can produce the qualitative synthetic data needed to train a standard translation model. This work proposes a self-training strategy where the output of the backward model is used to improve the model itself through the forward translation technique. The technique was shown to improve baseline low resource IWSLT'14 English-German and IWSLT'15 English-Vietnamese backward translation models by 11.06 and 1.5 BLEUs respectively. The synthetic data generated...

Research paper thumbnail of Iterative Batch Back-Translation for Neural Machine Translation: A Conceptual Model

ArXiv, 2020

An effective method to generate a large number of parallel sentences for training improved neural... more An effective method to generate a large number of parallel sentences for training improved neural machine translation (NMT) systems is the use of back-translations of the target-side monolingual data. Recently, iterative back-translation has been shown to outperform standard back-translation albeit on some language pairs. This work proposes the iterative batch back-translation that is aimed at enhancing the standard iterative back-translation and enabling the efficient utilization of more monolingual data. After each iteration, improved back-translations of new sentences are added to the parallel data that will be used to train the final forward model. The work presents a conceptual model of the proposed approach.

Research paper thumbnail of A Hybrid Approach for Improved Low Resource Neural Machine Translation using Monolingual Data

arXiv: Computation and Language, 2020

Many language pairs are low resource, meaning the amount and/or quality of available parallel dat... more Many language pairs are low resource, meaning the amount and/or quality of available parallel data is not sufficient to train a neural machine translation (NMT) model which can reach an acceptable standard of accuracy. Many works have explored using the readily available monolingual data in either or both of the languages to improve the standard of translation models in low, and even high, resource languages. One of the most successful of such works is the back-translation that utilizes the translations of the target language monolingual data to increase the amount of the training data. The quality of the backward model which is trained on the available parallel data has been shown to determine the performance of the back-translation approach. Despite this, only the forward model is improved on the monolingual target data in standard back-translation. A previous study proposed an iterative back-translation approach for improving both models over several iterations. But unlike in the...

Research paper thumbnail of Design and Comparison Migration between Ipv4 and Ipv6 Transition Techniques

IPv4 which is the old version of Internet Protocol has a new successor named IP Next Generation (... more IPv4 which is the old version of Internet Protocol has a new successor named IP Next Generation (IPng) or IPv6 developed by Internet Engineering Task Force (IETF). This new version is developed specifically to resolve some issues of IPv4 like scalability, performance and reliability. Although new version is ready for usage, it is obvious that it will take years to transit fully from IPv4 to IPv6. We have to use these two versions together for a long time. Therefore, we have to investigate and know transition mechanisms that we can use during transition period to achieve a transition with minimum problem. This research defines the essential information about compatibility transition mechanisms between IPv4-IPv6. Dual Stack is one of the IPv4-IPv6 compatible mechanism by running both IPv4 stack and IPv6 stack in a single node. 6 to 4 tunneling mechanism encrypts IPv6 packets in IPv4 packets to make communications possible, from IPv6 network over IPv4 network. This has been configured ...

Research paper thumbnail of Enhanced Back-Translation for Low Resource Neural Machine Translation Using Self-training

Communications in Computer and Information Science

Improving neural machine translation (NMT) models using the back-translations of the monolingual ... more Improving neural machine translation (NMT) models using the back-translations of the monolingual target data (synthetic parallel data) is currently the state-of-the-art approach for training improved translation systems. The quality of the backward system-which is trained on the available parallel data and used for the back-translation-has been shown in many studies to affect the performance of the final NMT model. In low resource conditions, the available parallel data is usually not enough to train a backward model that can produce the qualitative synthetic data needed to train a standard translation model. This work proposes a self-training strategy where the output of the backward model is used to improve the model itself through the forward translation technique. The technique was shown to improve baseline low resource IWSLT'14 English-German and IWSLT'15 English-Vietnamese backward translation models by 11.06 and 1.5 BLEUs respectively. The synthetic data generated by the improved English-German backward model was used to train a forward model which out-performed another forward model trained using standard back-translation by 2.7 BLEU.

Research paper thumbnail of Using Self-Training to Improve Back-Translation in Low Resource Neural Machine Translation

ArXiv, 2020

Improving neural machine translation (NMT) models using the back-translations of the monolingual ... more Improving neural machine translation (NMT) models using the back-translations of the monolingual target data (synthetic parallel data) is currently the state-of-the-art approach for training improved translation systems. The quality of the backward system - which is trained on the available parallel data and used for the back-translation - has been shown in many studies to affect the performance of the final NMT model. In low resource conditions, the available parallel data is usually not enough to train a backward model that can produce the qualitative synthetic data needed to train a standard translation model. This work proposes a self-training strategy where the output of the backward model is used to improve the model itself through the forward translation technique. The technique was shown to improve baseline low resource IWSLT'14 English-German and IWSLT'15 English-Vietnamese backward translation models by 11.06 and 1.5 BLEUs respectively. The synthetic data generated...

Research paper thumbnail of Implementation of Microcontroller Based Distance Relay

European Journal of Electrical Engineering and Computer Science

In this project microcontroller 8051 was used for comparisons between set Impedance and calculate... more In this project microcontroller 8051 was used for comparisons between set Impedance and calculated Impedance for based distance relay of the protection of transmission line, based on that trip signal is send to circuit breaker. The Hardware consisted of main CT and VT, auxiliary VT and CT ,BLF ,sample and hold(S/H) circuit ,multiplexer ,ADC and microcontroller 8051.In which the main CT and VT step down the current and voltage respectably passed to auxiliary CT and VT for protecting the remaining components and convert the current into voltage in the output of auxiliary CT, then passed it to band limit filter to limit the frequency for sampling hold after sampling and hold circuit the signal is passed to multiplexer then to the ADC for analogue to digital conversion, after the conversion the digital signals is passed to the 8051 microcontroller to processing the signals that is calculating the average, impedance and compare it with the set impedance (store impedance) to generate tr...