Professor Mo Saraee - Academia.edu (original) (raw)
Papers by Professor Mo Saraee
2022 International Conference on Electrical Engineering and Informatics (ICELTICs)
Welcome to the Book of Abstracts for the 2018 SPARC conference. This year we not only celebrate t... more Welcome to the Book of Abstracts for the 2018 SPARC conference. This year we not only celebrate the work of our PGRs but also the launch of our Doctoral School, which makes this year’s conference extra special. Once again we have received a tremendous contribution from our postgraduate research community; with over 100 presenters, the conference truly showcases a vibrant PGR community at Salford. These abstracts provide a taster of the research strengths of their works, and provide delegates with a reference point for networking and initiating critical debate. With such wide-ranging topics being showcased, we encourage you to take up this great opportunity to engage with researchers working in different subject areas from your own. To meet global challenges, high impact research inevitably requires interdisciplinary collaboration. This is recognised by all major research funders. Therefore engaging with the work of others and forging collaborations across subject areas is an essenti...
Angewandte Chemie, 2019
Angabe der unten stehenden Digitalobjekt-Identifizierungsnummer (DOI) zitiert werden. Die deutsch... more Angabe der unten stehenden Digitalobjekt-Identifizierungsnummer (DOI) zitiert werden. Die deutsche Übersetzung wird gemeinsam mit der endgültigen englischen Fassung erscheinen. Die endgültige englische Fassung (Version of Record) wird ehestmöglich nach dem Redigieren und einem Korrekturgang als Early-View-Beitrag erscheinen und kann sich naturgemäß von der AA-Fassung unterscheiden. Leser sollten daher die endgültige Fassung, sobald sie veröffentlicht ist, verwenden. Für die AA-Fassung trägt der Autor die alleinige Verantwortung.
Proceedings of the 2019 2nd International Conference on Geoinformatics and Data Analysis - ICGDA 2019, 2019
Diabetes is a pandemic that is growing globally, and by the year 2030 it is expected to effect th... more Diabetes is a pandemic that is growing globally, and by the year 2030 it is expected to effect three people every 10 minutes. In the UK, it is estimated that by 2025, 5 million people will have diabetes. Diabetes is currently costing the British National Health Service (NHS) over £1.5m an hour. This equates to 10% of the NHS budget for England and Wales or over £25,000 being spent on diabetes every minute. It is important to minimise these costs by employing new techniques to curtail existing cases and limit new cases. This goal will be accomplished through patient education and self-management. Self-management and selfmonitoring play a significant role in restraining diabetes complications. A major component of self-management involves regular blood testing and detailed record keeping through comparative analysis, extensive surveying of diabetic's patients and medical professionals were carried out. The outcome of this research was used to determine how to best create a global and easily accessible diabetes management etoolkit, that will help the diabetic community alongside the medical community to reduce the harmful effects of this disease. The e-toolkit, that has been formed successfully, can connect the patient to their doctor. It will provide the patient with a single means of recording every important vital health factor and simultaneously allowing the doctor to access real-time monitoring. The e-toolkit proposed in this paper would facilitate in bringing both the medical and diabetic group of people closer together resulting in a strengthened relationship. Patients will be able to record each imperative health factor whilst having the ability to communicate with their doctor and in turn, effectively managing their diabetes to their utmost potential.
In this paper a multi agent model for predicting monthly maximum and minimum quote prices has bee... more In this paper a multi agent model for predicting monthly maximum and minimum quote prices has been proposed. This model is based on the training of Elman neural networks and using particle swarm optimization for obtaining the best parameters of the neural networks. Also one method for reducing the effects of overfitting problem is suggested. This method averages the outputs of an ensemble network, given noisy data as inout, to predict the final results.. Finally the results of using this model on a sample data set are presented and the effectiveness of this model is illustrated.
ACM Computing Surveys, 2012
Ontologies are often viewed as the answer to the need for interoperable semantics in modern infor... more Ontologies are often viewed as the answer to the need for interoperable semantics in modern information systems. The explosion of textual information on the Read/Write Web coupled with the increasing demand for ontologies to power the Semantic Web have made (semi-)automatic ontology learning from text a very promising research area. This together with the advanced state in related areas, such as natural language processing, have fueled research into ontology learning over the past decade. This survey looks at how far we have come since the turn of the millennium and discusses the remaining challenges that will define the research directions in this area in the near future.
This paper presents an approach to the shortest path routing problem that uses one of the most po... more This paper presents an approach to the shortest path routing problem that uses one of the most popular learning algorithms. The Genetic Algorithm (GA) is one of the most powerful and successful method in stochastic search and optimization techniques based on the principles of the evolution theory. The crossover operation examines the current solutions in order to find better ones and the mutation operation introduces a new alternative route. The shortest path problem concentrates on finding the path with minimum distance, time or cost from a source node to the goal node. Routing decisions are based on constantly changing predictions of the weights. Finally we arrange some experiments to testify the efficiency of our method. In most of the experiments, the Genetic algorithms found the shortest path in a quick time and had good performance.
Proceedings of the International Conference on Machine Learning; Models, Technologies and Applications, 2004
Supervised text categorisation is a significant tool considering the vast amount of structured, u... more Supervised text categorisation is a significant tool considering the vast amount of structured, unstruc-tured, or semi-structured texts that are available from internal or external enterprise resources. The goal of supervised text categorisation is to assign text documents to finite pre-specified categories in order to extract and automatically organise information coming from these resources. This paper pro-poses the implementation of a generic application–SVM Categorizer using the Support Vector Ma-chines algorithm with an innovative statistical ...
Proceedings of Panhellenic Conference of New Information Technology, Athens, Greece, Oct 8, 1998
In this paper we describe our approaches to data mining in temporal databases by introducing Easy... more In this paper we describe our approaches to data mining in temporal databases by introducing Easy Miner, our data mining system developed at UMIST. Easy Miner integrates machine learning methodologies with database technologies and efficiently and effectively extract interesting rules from data. The discovery components of Easy Miner, relevance analysis, classification and association rules finder are presented with the algorithms behind each component. We also show the effectiveness of these algorithms ...
CO, NOx, HC and Particle mass have been monitored in different vehicle emission standards and Par... more CO, NOx, HC and Particle mass have been monitored in different vehicle emission standards and Particle number (PN) has been added to standards recently. The EU has proposed a solid particle PN limit in Euro 5b and Euro 6. The PN limit for low duty vehicles (LDVs) is set to be 6.0 × 1011 (#/km). Within the recent decades of the European Union, there has been an overall reduction in emissions of air pollutants. It can however be reported that emissions of diesel engines that are measured within laboratory conditions are different from what is happening in real world driving condition. To combat this issue, a more realistic measurement method called Portable Emission Measurement System (PEMS) is recommended. There are very low amounts of real world emission data from PEMS for passenger cars [1]. Also, the available reports mostly concentrated on gaseous emissions such as NOx [2] and particle emissions from diesel passenger cars in real world is rarely studied. The aim of this research...
International Symposium on Neural Networks, 2008
In this paper a new method to resolve the overfitting problem for predicting complex systemspsila... more In this paper a new method to resolve the overfitting problem for predicting complex systemspsila behavior has been proposed. This problem occurs when a neural network loses its generalization. The method is based on the training of recurrent neural networks and using simulated annealing for the optimization of their generalization. The major work is done based on the idea of
2019 IEEE Conference on Sustainable Utilization and Development in Engineering and Technologies (CSUDET)
The European Commission estimates that around 135,000 people are seriously injured on Europe's ro... more The European Commission estimates that around 135,000 people are seriously injured on Europe's roads each year. The road traffic injuries are a significant but neglected global general public health problem, needing rigorous attempts for effective and workable prevention. One of the ways to decrease the amount of traffic accidents is to conduct an indepth assessment on the historically documented road traffic incident data and understand the cause of the accidents and factors associated with incident severity. It may provide crucial information for emergency services to evaluate the severity level of accidents, estimate the potential impacts of the casualties, and ultimately it might help to improve the road safety. In this study author is trying to identify the factors that correlate with the slight and serious (including fatal) Road Traffic Accident using Decision Tree classification algorithms using UK STATS19 dataset. Also, author is exploring the possibility of enhancing the knowledge gain from Decision Tree classification algorithms using Time-Series Calendar Heatmap in order to identify hidden temporal patterns. The methodology described in this study offers significant advantages over understanding correlation between hour and month of the accident and the severity of the accident. Although this study is based on a region in North of England, the approach can be applicable to other areas in UK and globally with similar kind of road side accident data. This study found out that combining classification methods like decision tree and time-series calendar heatmaps cam be a useful tool for accurately classifying roadside traffic accidents according to their injury severity.
2021 International Symposium on Electrical, Electronics and Information Engineering, 2021
Chatbots are computer programs designed to simulate conversation by interacting with a human user... more Chatbots are computer programs designed to simulate conversation by interacting with a human user. In this paper we present a chatbot framework designed specifically to aid prolonged grief disorder (PGD) sufferers by replicating the techniques performed during cold readings. Our initial framework performed an association rule analysis on transcripts of real-world cold reading performances, in order to generate the required data as used in traditional rules based chatbots. However due to the structure of cold readings the traditional approach was unable to determine a satisfactory set of rules. Therefore, in this paper we discuss the limitations of this approach and subsequently provide a generative solution using sequence-to-sequence modeling with long short-term memory. We demonstrate how our generative chatbot is therefore able to provide appropriate responses to the majority of inputs. However, as inappropriate responses can present a risk to sensitive PGD sufferers we suggest a ...
2019 IEEE Conference on Sustainable Utilization and Development in Engineering and Technologies (CSUDET), 2019
In national and regional level, understanding of factors associated with public health issues lik... more In national and regional level, understanding of factors associated with public health issues like mental health is paramount important to improve the awareness. This study aims to use the data mining techniques such as association rule mining to improve the degree of understanding the mental health among various geographical areas by identifying various vital behavioural factors associated with mental health issues. The study will produce interesting relationships among the behavioural factors in form of Association rules. The outcomes of this research will be beneficial to organisations that work in public health sector to improve mental health among the citizens. Also, this proposed new data science approach will be beneficial to improve the degree of understanding by identifying factors associated with mental health within the different geographical areas such as city or state level. The study found that states in US which have low excessive drinking percentage and high obesity ...
In order to succeed in the global competition, organizations need to understand and monitor the r... more In order to succeed in the global competition, organizations need to understand and monitor the rate of data influx. The acquisition of continuous data has been extremely outstretched as a concern in many fields. Recently, frequent patterns in data streams have been a challenging task in the field of data mining and knowledge discovery. Most of these datasets generated are in the form of a stream (stream data), thereby posing a challenge of being continuous. Therefore, the process of extracting knowledge structures from continuous rapid data records is termed as stream mining. This study conceptualizes the process of detecting outliers and responding to stream data. This is done by proposing a Compressed Stream Pattern algorithm, which dynamically generates a frequency descending prefix tree structure with only a singlepass over the data. We show that applying tree restructuring techniques can considerably minimize the mining time on various datasets. KeywordsData Mining; Frequent P...
Proceedings of the International Conference on Geoinformatics and Data Analysis, 2018
Geospatial Big Data analytics are changing the way that businesses operate in many industries. Al... more Geospatial Big Data analytics are changing the way that businesses operate in many industries. Although a good number of research works have reported in the literature on geospatial data analytics and real-time data processing of large spatial data streams, only a few have addressed the full geospatial big data analytics project lifecycle and geospatial data science project lifecycle. Big data analysis differs from traditional data analysis primarily due to the volume, velocity and variety characteristics of the data being processed. One of a motivation of introducing new framework is to address these big data analysis challenges. Geospatial data science projects differ from most traditional data analysis projects because they could be complex and in need of advanced technologies in comparison to the traditional data analysis projects. For this reason, it is essential to have a process to govern the project and ensure that the project participants are competent enough to carry on th...
2019 IEEE Conference on Sustainable Utilization and Development in Engineering and Technologies (CSUDET), 2019
The European Commission estimates that around 135,000 people are seriously injured on Europe'... more The European Commission estimates that around 135,000 people are seriously injured on Europe's roads each year. The road traffic injuries are a significant but neglected global general public health problem, needing rigorous attempts for effective and workable prevention. One of the ways to decrease the amount of traffic accidents is to conduct an in-depth assessment on the historically documented road traffic incident data and understand the cause of the accidents and factors associated with incident severity. It may provide crucial information for emergency services to evaluate the severity level of accidents, estimate the potential impacts of the casualties, and ultimately it might help to improve the road safety. In this study author is trying to identify the factors that correlate with the slight and serious (including fatal) Road Traffic Accident using Decision Tree classification algorithms using UK STATS19 dataset. Also, author is exploring the possibility of enhancing t...
2019 IEEE Symposium on Computers and Communications (ISCC), Jun 1, 2019
Understanding public mental health issues and finding solutions can be complex and requires advan... more Understanding public mental health issues and finding solutions can be complex and requires advanced techniques, compared to conventional data analysis projects. It is important to have a comprehensive project management process to ensure that project associates are competent and have enough knowledge to implement the process. Therefore, this paper presents a new framework that mental health professionals can use to solve challenges they face. Although a large number of research papers have been published on public mental health, few have addressed the use of data science in public mental health. Recently, Data Science has changed the way we manage, analyze and leverage data in healthcare industry. Data science projects differ from conventional data analysis, primarily because of the scientific approach used during data science projects. One of the motives for introducing a new framework is to motivate healthcare professionals to use "Data Science" to address the challenges of mental health. Having a good data analysis framework and clear guidelines for a comprehensive analysis is always a plus point. It also helps to predict the time and resources needed in the early in the process to get a clear idea of the problem to be solved.
Journal of Intelligent & Fuzzy Systems
ABSTRACT Data mining techniques are often confined to the delivery of frequent patterns and stop ... more ABSTRACT Data mining techniques are often confined to the delivery of frequent patterns and stop short of suggesting how to act on these patterns for business decision-making. They require human experts to post-process the discovered patterns manually. Therefore a significant need exists for techniques and tools with the ability to assist users in analyzing a large number of patterns to find usable knowledge. Action mining is one of these techniques which intelligently and automatically suggests some changes in the state of an object with the aim of gaining some profit in the corresponding domain. Up to now little research has been done in this field; in all cases continuous-valued data is handled by discretizing the associated attributes in advance or during the learning process. One inherent disadvantage in these methods is that using this sharp behavior can result in missing the optimal action. To overcome this problem this paper presents a method based on fuzzy set theory. In this paper, we concentrate on the fuzzy set based approach for the enhancement of Yang's method and present an algorithm that suggests actions which will decrease the degree to which a certain object belongs to an undesired status and increase the degree to which it belongs to a desired one. Our algorithm takes into account the fuzzy cost of actions, and further, it attempts to maximize the fuzzy net profit. The contribution of the work is in taking the output from fuzzy decision trees, and producing novel, actionable knowledge through automatic fuzzy post-processing. The performance of the proposed algorithm is compared with Yang's method using several real-life datasets taken from the UCI Machine Learning Repository. Experimental results show that the proposed algorithm outperforms Yang's method not only in finding more actions but also in finding actions with more fuzzy net profit.
Swarm and Evolutionary Computation
Online social networks play an important role in marketing services. Influence maximization is a ... more Online social networks play an important role in marketing services. Influence maximization is a major challenge given its goal to find the most influential users in a social network. Increasing the number of influenced users at the end of a diffusion process while decreasing the time of diffusion are two main objectives of the influence maximization problem. The goal of this paper is to find multiple sets of influential users such that each of them is the best set to spread influence for a specific time bound. Considering two adverse objectives, increasing influence and decreasing diffusion time, we employ the NSGA-II algorithm which is a powerful algorithm in multi-objective optimization to find different seed sets with high influence at different diffusion times. Since social networks are large, computing influence and diffusion time of all chromosomes in each iteration will be challenging and computationally expensive. Therefore, we propose two methods which can estimate the expected influence and diffusion time of a seed set in an efficient manner. Providing the set of all potentially optimal solutions, help a decision maker evaluate the tradeoffs between the two objectives, i.e., the number of influenced users and diffusion time. In addition, we develop an approach for selecting seed sets, which have optimal influence for specific time bounds, from the resulting Pareto front of the NSGA-II. Finally, we show that applying our algorithm to real social networks outperforms existing algorithms for influence maximization problem. The results show a good compromise between two objectives and the final seed sets result in high influence for different time bounds.
2022 International Conference on Electrical Engineering and Informatics (ICELTICs)
Welcome to the Book of Abstracts for the 2018 SPARC conference. This year we not only celebrate t... more Welcome to the Book of Abstracts for the 2018 SPARC conference. This year we not only celebrate the work of our PGRs but also the launch of our Doctoral School, which makes this year’s conference extra special. Once again we have received a tremendous contribution from our postgraduate research community; with over 100 presenters, the conference truly showcases a vibrant PGR community at Salford. These abstracts provide a taster of the research strengths of their works, and provide delegates with a reference point for networking and initiating critical debate. With such wide-ranging topics being showcased, we encourage you to take up this great opportunity to engage with researchers working in different subject areas from your own. To meet global challenges, high impact research inevitably requires interdisciplinary collaboration. This is recognised by all major research funders. Therefore engaging with the work of others and forging collaborations across subject areas is an essenti...
Angewandte Chemie, 2019
Angabe der unten stehenden Digitalobjekt-Identifizierungsnummer (DOI) zitiert werden. Die deutsch... more Angabe der unten stehenden Digitalobjekt-Identifizierungsnummer (DOI) zitiert werden. Die deutsche Übersetzung wird gemeinsam mit der endgültigen englischen Fassung erscheinen. Die endgültige englische Fassung (Version of Record) wird ehestmöglich nach dem Redigieren und einem Korrekturgang als Early-View-Beitrag erscheinen und kann sich naturgemäß von der AA-Fassung unterscheiden. Leser sollten daher die endgültige Fassung, sobald sie veröffentlicht ist, verwenden. Für die AA-Fassung trägt der Autor die alleinige Verantwortung.
Proceedings of the 2019 2nd International Conference on Geoinformatics and Data Analysis - ICGDA 2019, 2019
Diabetes is a pandemic that is growing globally, and by the year 2030 it is expected to effect th... more Diabetes is a pandemic that is growing globally, and by the year 2030 it is expected to effect three people every 10 minutes. In the UK, it is estimated that by 2025, 5 million people will have diabetes. Diabetes is currently costing the British National Health Service (NHS) over £1.5m an hour. This equates to 10% of the NHS budget for England and Wales or over £25,000 being spent on diabetes every minute. It is important to minimise these costs by employing new techniques to curtail existing cases and limit new cases. This goal will be accomplished through patient education and self-management. Self-management and selfmonitoring play a significant role in restraining diabetes complications. A major component of self-management involves regular blood testing and detailed record keeping through comparative analysis, extensive surveying of diabetic's patients and medical professionals were carried out. The outcome of this research was used to determine how to best create a global and easily accessible diabetes management etoolkit, that will help the diabetic community alongside the medical community to reduce the harmful effects of this disease. The e-toolkit, that has been formed successfully, can connect the patient to their doctor. It will provide the patient with a single means of recording every important vital health factor and simultaneously allowing the doctor to access real-time monitoring. The e-toolkit proposed in this paper would facilitate in bringing both the medical and diabetic group of people closer together resulting in a strengthened relationship. Patients will be able to record each imperative health factor whilst having the ability to communicate with their doctor and in turn, effectively managing their diabetes to their utmost potential.
In this paper a multi agent model for predicting monthly maximum and minimum quote prices has bee... more In this paper a multi agent model for predicting monthly maximum and minimum quote prices has been proposed. This model is based on the training of Elman neural networks and using particle swarm optimization for obtaining the best parameters of the neural networks. Also one method for reducing the effects of overfitting problem is suggested. This method averages the outputs of an ensemble network, given noisy data as inout, to predict the final results.. Finally the results of using this model on a sample data set are presented and the effectiveness of this model is illustrated.
ACM Computing Surveys, 2012
Ontologies are often viewed as the answer to the need for interoperable semantics in modern infor... more Ontologies are often viewed as the answer to the need for interoperable semantics in modern information systems. The explosion of textual information on the Read/Write Web coupled with the increasing demand for ontologies to power the Semantic Web have made (semi-)automatic ontology learning from text a very promising research area. This together with the advanced state in related areas, such as natural language processing, have fueled research into ontology learning over the past decade. This survey looks at how far we have come since the turn of the millennium and discusses the remaining challenges that will define the research directions in this area in the near future.
This paper presents an approach to the shortest path routing problem that uses one of the most po... more This paper presents an approach to the shortest path routing problem that uses one of the most popular learning algorithms. The Genetic Algorithm (GA) is one of the most powerful and successful method in stochastic search and optimization techniques based on the principles of the evolution theory. The crossover operation examines the current solutions in order to find better ones and the mutation operation introduces a new alternative route. The shortest path problem concentrates on finding the path with minimum distance, time or cost from a source node to the goal node. Routing decisions are based on constantly changing predictions of the weights. Finally we arrange some experiments to testify the efficiency of our method. In most of the experiments, the Genetic algorithms found the shortest path in a quick time and had good performance.
Proceedings of the International Conference on Machine Learning; Models, Technologies and Applications, 2004
Supervised text categorisation is a significant tool considering the vast amount of structured, u... more Supervised text categorisation is a significant tool considering the vast amount of structured, unstruc-tured, or semi-structured texts that are available from internal or external enterprise resources. The goal of supervised text categorisation is to assign text documents to finite pre-specified categories in order to extract and automatically organise information coming from these resources. This paper pro-poses the implementation of a generic application–SVM Categorizer using the Support Vector Ma-chines algorithm with an innovative statistical ...
Proceedings of Panhellenic Conference of New Information Technology, Athens, Greece, Oct 8, 1998
In this paper we describe our approaches to data mining in temporal databases by introducing Easy... more In this paper we describe our approaches to data mining in temporal databases by introducing Easy Miner, our data mining system developed at UMIST. Easy Miner integrates machine learning methodologies with database technologies and efficiently and effectively extract interesting rules from data. The discovery components of Easy Miner, relevance analysis, classification and association rules finder are presented with the algorithms behind each component. We also show the effectiveness of these algorithms ...
CO, NOx, HC and Particle mass have been monitored in different vehicle emission standards and Par... more CO, NOx, HC and Particle mass have been monitored in different vehicle emission standards and Particle number (PN) has been added to standards recently. The EU has proposed a solid particle PN limit in Euro 5b and Euro 6. The PN limit for low duty vehicles (LDVs) is set to be 6.0 × 1011 (#/km). Within the recent decades of the European Union, there has been an overall reduction in emissions of air pollutants. It can however be reported that emissions of diesel engines that are measured within laboratory conditions are different from what is happening in real world driving condition. To combat this issue, a more realistic measurement method called Portable Emission Measurement System (PEMS) is recommended. There are very low amounts of real world emission data from PEMS for passenger cars [1]. Also, the available reports mostly concentrated on gaseous emissions such as NOx [2] and particle emissions from diesel passenger cars in real world is rarely studied. The aim of this research...
International Symposium on Neural Networks, 2008
In this paper a new method to resolve the overfitting problem for predicting complex systemspsila... more In this paper a new method to resolve the overfitting problem for predicting complex systemspsila behavior has been proposed. This problem occurs when a neural network loses its generalization. The method is based on the training of recurrent neural networks and using simulated annealing for the optimization of their generalization. The major work is done based on the idea of
2019 IEEE Conference on Sustainable Utilization and Development in Engineering and Technologies (CSUDET)
The European Commission estimates that around 135,000 people are seriously injured on Europe's ro... more The European Commission estimates that around 135,000 people are seriously injured on Europe's roads each year. The road traffic injuries are a significant but neglected global general public health problem, needing rigorous attempts for effective and workable prevention. One of the ways to decrease the amount of traffic accidents is to conduct an indepth assessment on the historically documented road traffic incident data and understand the cause of the accidents and factors associated with incident severity. It may provide crucial information for emergency services to evaluate the severity level of accidents, estimate the potential impacts of the casualties, and ultimately it might help to improve the road safety. In this study author is trying to identify the factors that correlate with the slight and serious (including fatal) Road Traffic Accident using Decision Tree classification algorithms using UK STATS19 dataset. Also, author is exploring the possibility of enhancing the knowledge gain from Decision Tree classification algorithms using Time-Series Calendar Heatmap in order to identify hidden temporal patterns. The methodology described in this study offers significant advantages over understanding correlation between hour and month of the accident and the severity of the accident. Although this study is based on a region in North of England, the approach can be applicable to other areas in UK and globally with similar kind of road side accident data. This study found out that combining classification methods like decision tree and time-series calendar heatmaps cam be a useful tool for accurately classifying roadside traffic accidents according to their injury severity.
2021 International Symposium on Electrical, Electronics and Information Engineering, 2021
Chatbots are computer programs designed to simulate conversation by interacting with a human user... more Chatbots are computer programs designed to simulate conversation by interacting with a human user. In this paper we present a chatbot framework designed specifically to aid prolonged grief disorder (PGD) sufferers by replicating the techniques performed during cold readings. Our initial framework performed an association rule analysis on transcripts of real-world cold reading performances, in order to generate the required data as used in traditional rules based chatbots. However due to the structure of cold readings the traditional approach was unable to determine a satisfactory set of rules. Therefore, in this paper we discuss the limitations of this approach and subsequently provide a generative solution using sequence-to-sequence modeling with long short-term memory. We demonstrate how our generative chatbot is therefore able to provide appropriate responses to the majority of inputs. However, as inappropriate responses can present a risk to sensitive PGD sufferers we suggest a ...
2019 IEEE Conference on Sustainable Utilization and Development in Engineering and Technologies (CSUDET), 2019
In national and regional level, understanding of factors associated with public health issues lik... more In national and regional level, understanding of factors associated with public health issues like mental health is paramount important to improve the awareness. This study aims to use the data mining techniques such as association rule mining to improve the degree of understanding the mental health among various geographical areas by identifying various vital behavioural factors associated with mental health issues. The study will produce interesting relationships among the behavioural factors in form of Association rules. The outcomes of this research will be beneficial to organisations that work in public health sector to improve mental health among the citizens. Also, this proposed new data science approach will be beneficial to improve the degree of understanding by identifying factors associated with mental health within the different geographical areas such as city or state level. The study found that states in US which have low excessive drinking percentage and high obesity ...
In order to succeed in the global competition, organizations need to understand and monitor the r... more In order to succeed in the global competition, organizations need to understand and monitor the rate of data influx. The acquisition of continuous data has been extremely outstretched as a concern in many fields. Recently, frequent patterns in data streams have been a challenging task in the field of data mining and knowledge discovery. Most of these datasets generated are in the form of a stream (stream data), thereby posing a challenge of being continuous. Therefore, the process of extracting knowledge structures from continuous rapid data records is termed as stream mining. This study conceptualizes the process of detecting outliers and responding to stream data. This is done by proposing a Compressed Stream Pattern algorithm, which dynamically generates a frequency descending prefix tree structure with only a singlepass over the data. We show that applying tree restructuring techniques can considerably minimize the mining time on various datasets. KeywordsData Mining; Frequent P...
Proceedings of the International Conference on Geoinformatics and Data Analysis, 2018
Geospatial Big Data analytics are changing the way that businesses operate in many industries. Al... more Geospatial Big Data analytics are changing the way that businesses operate in many industries. Although a good number of research works have reported in the literature on geospatial data analytics and real-time data processing of large spatial data streams, only a few have addressed the full geospatial big data analytics project lifecycle and geospatial data science project lifecycle. Big data analysis differs from traditional data analysis primarily due to the volume, velocity and variety characteristics of the data being processed. One of a motivation of introducing new framework is to address these big data analysis challenges. Geospatial data science projects differ from most traditional data analysis projects because they could be complex and in need of advanced technologies in comparison to the traditional data analysis projects. For this reason, it is essential to have a process to govern the project and ensure that the project participants are competent enough to carry on th...
2019 IEEE Conference on Sustainable Utilization and Development in Engineering and Technologies (CSUDET), 2019
The European Commission estimates that around 135,000 people are seriously injured on Europe'... more The European Commission estimates that around 135,000 people are seriously injured on Europe's roads each year. The road traffic injuries are a significant but neglected global general public health problem, needing rigorous attempts for effective and workable prevention. One of the ways to decrease the amount of traffic accidents is to conduct an in-depth assessment on the historically documented road traffic incident data and understand the cause of the accidents and factors associated with incident severity. It may provide crucial information for emergency services to evaluate the severity level of accidents, estimate the potential impacts of the casualties, and ultimately it might help to improve the road safety. In this study author is trying to identify the factors that correlate with the slight and serious (including fatal) Road Traffic Accident using Decision Tree classification algorithms using UK STATS19 dataset. Also, author is exploring the possibility of enhancing t...
2019 IEEE Symposium on Computers and Communications (ISCC), Jun 1, 2019
Understanding public mental health issues and finding solutions can be complex and requires advan... more Understanding public mental health issues and finding solutions can be complex and requires advanced techniques, compared to conventional data analysis projects. It is important to have a comprehensive project management process to ensure that project associates are competent and have enough knowledge to implement the process. Therefore, this paper presents a new framework that mental health professionals can use to solve challenges they face. Although a large number of research papers have been published on public mental health, few have addressed the use of data science in public mental health. Recently, Data Science has changed the way we manage, analyze and leverage data in healthcare industry. Data science projects differ from conventional data analysis, primarily because of the scientific approach used during data science projects. One of the motives for introducing a new framework is to motivate healthcare professionals to use "Data Science" to address the challenges of mental health. Having a good data analysis framework and clear guidelines for a comprehensive analysis is always a plus point. It also helps to predict the time and resources needed in the early in the process to get a clear idea of the problem to be solved.
Journal of Intelligent & Fuzzy Systems
ABSTRACT Data mining techniques are often confined to the delivery of frequent patterns and stop ... more ABSTRACT Data mining techniques are often confined to the delivery of frequent patterns and stop short of suggesting how to act on these patterns for business decision-making. They require human experts to post-process the discovered patterns manually. Therefore a significant need exists for techniques and tools with the ability to assist users in analyzing a large number of patterns to find usable knowledge. Action mining is one of these techniques which intelligently and automatically suggests some changes in the state of an object with the aim of gaining some profit in the corresponding domain. Up to now little research has been done in this field; in all cases continuous-valued data is handled by discretizing the associated attributes in advance or during the learning process. One inherent disadvantage in these methods is that using this sharp behavior can result in missing the optimal action. To overcome this problem this paper presents a method based on fuzzy set theory. In this paper, we concentrate on the fuzzy set based approach for the enhancement of Yang's method and present an algorithm that suggests actions which will decrease the degree to which a certain object belongs to an undesired status and increase the degree to which it belongs to a desired one. Our algorithm takes into account the fuzzy cost of actions, and further, it attempts to maximize the fuzzy net profit. The contribution of the work is in taking the output from fuzzy decision trees, and producing novel, actionable knowledge through automatic fuzzy post-processing. The performance of the proposed algorithm is compared with Yang's method using several real-life datasets taken from the UCI Machine Learning Repository. Experimental results show that the proposed algorithm outperforms Yang's method not only in finding more actions but also in finding actions with more fuzzy net profit.
Swarm and Evolutionary Computation
Online social networks play an important role in marketing services. Influence maximization is a ... more Online social networks play an important role in marketing services. Influence maximization is a major challenge given its goal to find the most influential users in a social network. Increasing the number of influenced users at the end of a diffusion process while decreasing the time of diffusion are two main objectives of the influence maximization problem. The goal of this paper is to find multiple sets of influential users such that each of them is the best set to spread influence for a specific time bound. Considering two adverse objectives, increasing influence and decreasing diffusion time, we employ the NSGA-II algorithm which is a powerful algorithm in multi-objective optimization to find different seed sets with high influence at different diffusion times. Since social networks are large, computing influence and diffusion time of all chromosomes in each iteration will be challenging and computationally expensive. Therefore, we propose two methods which can estimate the expected influence and diffusion time of a seed set in an efficient manner. Providing the set of all potentially optimal solutions, help a decision maker evaluate the tradeoffs between the two objectives, i.e., the number of influenced users and diffusion time. In addition, we develop an approach for selecting seed sets, which have optimal influence for specific time bounds, from the resulting Pareto front of the NSGA-II. Finally, we show that applying our algorithm to real social networks outperforms existing algorithms for influence maximization problem. The results show a good compromise between two objectives and the final seed sets result in high influence for different time bounds.