Pre-trained language models with limited data for intent classification
- Authors: Kasthuriarachchy, Buddhika , Chetty, Madhu , Karmakar, Gour , Walls, Darren
- Date: 2020
- Type: Text , Conference proceedings , Conference paper
- Relation: 2020 International Joint Conference on Neural Networks, IJCNN 2020
- Full Text: false
- Reviewed:
- Description: Intent analysis is capturing the attention of both the industry and academia due to its commercial and noncommercial significance. The rapid growth of unstructured data of micro-blogging platforms, such as Twitter and Facebook, are amongst the important sources for intent analysis. However, the social media data are often noisy and diverse, thus making the task very challenging. Further, the intent analysis frequently suffers from lack of sufficient data because the labeled datasets are often manually annotated. Recently, BERT (Bidirectional Encoder Representation from Transformers), a state-of-the-art language representation model, has attracted attention for accurate language modelling. In this paper, we investigate the application of BERT for its suitability for intent analysis. We study the fine-tuning of the BERT model through inductive transfer learning and investigate methods to overcome the challenges due to limited data availability by proposing a novel semantic data augmentation approach. This technique generates synthetic sentences while preserving the label-compatibility using the semantic meaning of the sentences, to improve the intent classification accuracy. Thus, based on the considerations for finetuning and data augmentation, a systematic and novel step-bystep methodology is presented for applying the linguistic model BERT for intent classification with limited data available. Our results show that the pre-trained language can be effectively used with noisy social media data to achieve state-of-the-art accuracy in intent analysis under low labeled-data regime. Moreover, our results also confirm that the proposed text augmentation technique is effective in eliminating noisy synthetic sentences, thereby achieving further performance improvements. © 2020 IEEE.
Assessing transformer oil quality using deep convolutional networks
- Authors: Alam, Mohammad , Karmakar, Gour , Islam, Syed , Kamruzzaman, Joarder , Chetty, Madhu , Lim, Suryani , Appuhamillage, Gayan , Chattopadhyay, Gopi , Wilcox, Steve , Verheyen, Vincent
- Date: 2019
- Type: Text , Conference proceedings , Conference paper
- Relation: 29th Australasian Universities Power Engineering Conference, AUPEC 2019
- Full Text:
- Reviewed:
- Description: Electrical power grids comprise a significantly large number of transformers that interconnect power generation, transmission and distribution. These transformers having different MVA ratings are critical assets that require proper maintenance to provide long and uninterrupted electrical service. The mineral oil, an essential component of any transformer, not only provides cooling but also acts as an insulating medium within the transformer. The quality and the key dissolved properties of insulating mineral oil for the transformer are critical with its proper and reliable operation. However, traditional chemical diagnostic methods are expensive and time-consuming. A transformer oil image analysis approach, based on the entropy value of oil, which is inexpensive, effective and quick. However, the inability of entropy to estimate the vital transformer oil properties such as equivalent age, Neutralization Number (NN), dissipation factor (tanδ) and power factor (PF); and many intuitively derived constants usage limit its estimation accuracy. To address this issue, in this paper, we introduce an innovative transformer oil analysis using two deep convolutional learning techniques such as Convolutional Neural Network (ConvNet) and Residual Neural Network (ResNet). These two deep neural networks are chosen for this project as they have superior performance in computer vision. After estimating the equivalent aging year of transformer oil from its image by our proposed method, NN, tanδ and PF are computed using that estimated age. Our deep learning based techniques can accurately predict the transformer oil equivalent age, leading to calculate NN, tanδ and PF more accurately. The root means square error of estimated equivalent age produced by entropy, ConvNet and ResNet based methods are 0.718, 0.122 and 0.065, respectively. ConvNet and ResNet based methods have reduced the error of the oil age estimation by 83% and 91%, respectively compared to that of the entropy method. Our proposed oil image analysis can calculate the equivalent age that is very close to the actual age for all images used in the experiment. © 2019 IEEE.
- Description: E1
Measuring trustworthiness of IoT image sensor data using other sensors' complementary multimodal data
- Authors: Islam, Mohammad , Karmakar, Gour , Kamruzzaman, Joarder , Murshed, Manzur
- Date: 2019
- Type: Text , Conference proceedings , Conference paper
- Relation: 18th IEEE International Conference on Trust, Security and Privacy in Computing and Communications/13th IEEE International Conference on Big Data Science and Engineering, TrustCom/BigDataSE 2019 p. 775-780
- Full Text: false
- Reviewed:
- Description: Trust of image sensor data is becoming increasingly important as the Internet of Things (IoT) applications grow from home appliances to surveillance. Up to our knowledge, there exists only one work in literature that estimates trustworthiness of digital images applied to forensic applications, based on a machine learning technique. The efficacy of this technique is heavily dependent on availability of an appropriate training set and adequate variation of IoT sensor data with noise, interference and environmental condition, but availability of such data cannot be assured always. Therefore, to overcome this limitation, a robust method capable of estimating trustworthy measure with high accuracy is needed. Lowering cost of sensors allow many IoT applications to use multiple types of sensors to observe the same event. In such cases, complementary multimodal data of one sensor can be exploited to measure trust level of another sensor data. In this paper, for the first time, we introduce a completely new approach to estimate the trustworthiness of an image sensor data using another sensor's numerical data. We develop a theoretical model using the Dempster-Shafer theory (DST) framework. The efficacy of the proposed model in estimating trust level of an image sensor data is analyzed by observing a fire event using IoT image and temperature sensor data in a residential setup under different scenarios. The proposed model produces highly accurate trust level in all scenarios with authentic and forged image data. © 2019 IEEE.
- Description: E1
Trusted autonomous vehicle : measuring trust using on-board unit data
- Authors: Chowdhury, Abdullahi , Karmakar, Gour , Kamruzzaman, Joarder
- Date: 2019
- Type: Text , Conference proceedings , Conference paper
- Relation: 18th IEEE International Conference on Trust, Security and Privacy in Computing and Communications/13th IEEE International Conference on Big Data Science and Engineering, TrustCom/BigDataSE 2019 p. 787-792
- Full Text: false
- Reviewed:
- Description: Vehicular Ad-hoc Networks (VANETs) play an essential role in ensuring safe, reliable and faster transportation with the help of an Intelligent Transportation system. The trustworthiness of vehicles in VANETs is extremely important to ensure the authenticity of messages and traffic information transmitted in extremely dynamic topographical conditions where vehicles move at high speed. False or misleading information may cause substantial traffic congestions, road accidents and may even cost lives. Many approaches exist in literature to measure the trustworthiness of GPS data and messages of an Autonomous Vehicle (AV). To the best of our knowledge, they have not considered the trustworthiness of other On-Board Unit (OBU) components of an AV, along with GPS data and transmitted messages, though they have a substantial relevance in overall vehicle trust measurement. In this paper, we introduce a novel model to measure the overall trustworthiness of an AV considering four different OBU components additionally. The performance of the proposed method is evaluated with a traffic simulation model developed by Simulation of Urban Mobility (SUMO) using realistic traffic data and considering different levels of uncertainty. © 2019 IEEE.
- Description: E1
Cuboid colour image segmentation using intuitive distance measure
- Authors: Tania, Sheikh , Murshed, Manzur , Teng, Shyh , Karmakar, Gour
- Date: 2018
- Type: Text , Conference proceedings
- Relation: 2018 International Conference on Image and Vision Computing New Zealand, IVCNZ 2018; Auckland, New Zealand; 19th-21st November 2018 Vol. 2018-November, p. 1-6
- Full Text:
- Reviewed:
- Description: In this paper, an improved algorithm for cuboid image segmentation is proposed. To address the two main limitations of the recently proposed cuboid segmentation algorithm, the improved algorithm substitutes colour quantization in HCL colour space with infinity norm distance in RGB colour space along with a different way to impose area thresholding. We also propose a new metric to evaluate the quality of segmentation. Experimental results show that the proposed cuboid segmentation algorithm significantly outperforms the existing cuboid segmentation algorithm in terms of quality of segmentation.
- Description: International Conference Image and Vision Computing New Zealand
Detecting intrusion in the traffic signals of an intelligent traffic system
- Authors: Chowdhury, Abdullahi , Karmakar, Gour , Kamruzzaman, Joarder , Saha, Tapash
- Date: 2018
- Type: Text , Conference proceedings
- Relation: 20th International Conference on Information and Communications Security, ICICS 2018; Lille, France; 29th-31st October 2018; published in Lecure Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) Vol. 11149 LNCS, p. 696-707
- Full Text: false
- Reviewed:
- Description: Traffic systems and signals are used to improve traffic flow, reduce congestion, increase travel time consistency and ensure safety of road users. Malicious interruption or manipulation of traffic signals may cause disastrous instants including huge delays, financial loss and loss of lives. Intrusion into traffic signals by hackers can create such interruption whose consequences will only increase with the introduction of driverless vehicles. Recently, many traffic signals across the world are reported to have intruded, highlighting the importance of accurate detection. To reduce the impact of an intrusion, in this paper, we introduce an intrusion detection technique using the flow rate and phase time of a traffic signal as evidential information to detect the presence of an intrusion. The information received from flow rate and phase time are fused with the Dempster Shaffer (DS) theory. Historical data are used to create the probability mass functions for both flow rate and phase time. We also developed a simulation model using a traffic simulator, namely SUMO for many types of real traffic situations including intrusion. The performance of the proposed Intrusion Detection System (IDS) is appraised with normal traffic condition and induced intrusions. Simulated results show our proposed system can successfully detect intruded traffic signals from normal signals with significantly high accuracy (above 91%).
- Description: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Detecting splicing and copy-move attacks in color images
- Authors: Islam, Mohammad , Karmakar, Gour , Kamruzzaman, Joarder , Murshed, Manzur , Kahandawa, Gayan , Parvin, Nahida
- Date: 2018
- Type: Text , Conference proceedings
- Relation: 2018 International Conference on Digital Image Computing: Techniques and Applications, DICTA 2018; Canberra, Australia; 10th-13th December 2018 p. 1-7
- Full Text:
- Reviewed:
- Description: Image sensors are generating limitless digital images every day. Image forgery like splicing and copy-move are very common type of attacks that are easy to execute using sophisticated photo editing tools. As a result, digital forensics has attracted much attention to identify such tampering on digital images. In this paper, a passive (blind) image tampering identification method based on Discrete Cosine Transformation (DCT) and Local Binary Pattern (LBP) has been proposed. First, the chroma components of an image is divided into fixed sized non-overlapping blocks and 2D block DCT is applied to identify the changes due to forgery in local frequency distribution of the image. Then a texture descriptor, LBP is applied on the magnitude component of the 2D-DCT array to enhance the artifacts introduced by the tampering operation. The resulting LBP image is again divided into non-overlapping blocks. Finally, summations of corresponding inter-cell values of all the LBP blocks are computed and arranged as a feature vector. These features are fed into a Support Vector Machine (SVM) with Radial Basis Function (RBF) as kernel to distinguish forged images from authentic ones. The proposed method has been experimented extensively on three publicly available well-known image splicing and copy-move detection benchmark datasets of color images. Results demonstrate the superiority of the proposed method over recently proposed state-of-the-art approaches in terms of well accepted performance metrics such as accuracy, area under ROC curve and others.
- Description: 2018 International Conference on Digital Image Computing: Techniques and Applications, DICTA 2018
Influence of clustering on the opinion formation dynamics in online social networks
- Authors: Das, Rajkumar , Kamruzzaman, Joarder , Karmakar, Gour
- Date: 2018
- Type: Text , Conference proceedings
- Relation: 25th International Conference on Neural Information Processing, ICONIP 2018; Siem Reap; Cambodia; 13th-16th December 2018; published in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) Vol. 11306 LNCS, p. 144-155
- Full Text: false
- Reviewed:
- Description: With the advent of Online Social Networks (OSNs), opinion formation dynamics continuously evolves, mainly because of the widespread use of OSNs as a platform of social interactions and our growing exposure to others’ opinions instantly. When presented with neighbours’ opinions in OSNs, the natural clustering ability of human agents enables them to perceive the grouping of opinions formed in the neighbourhood. A group with similar opinions exhibits stronger influence on an agent than the individual group members. Distance-based opinion formation models only consider the influence of neighbours who are within a confidence bound threshold in the opinion space. However, a bigger group formed outside this distance threshold can exhibit stronger influence than a group within the bound, especially when that group contains influential or popular agents like leaders. To the knowledge of the authors, the proposed model is the first to consider the impact of clustering capability of agent and incorporates the influence of opinion clusters (groups) formed outside the confidence bound. Simulation results show that our model can capture several characteristics of real-world opinion dynamics. © Springer Nature Switzerland AG 2018.
Large scale modeling of genetic networks using gene knockout data
- Authors: Youseph, Ahammed , Chetty, Madhu , Karmakar, Gour
- Date: 2018
- Type: Text , Conference proceedings
- Relation: 2018 Australasian Computer Science Week Multiconference, ACSW 2018; Brisbane, Australia; 29th January-2nd February 2018; published in ACM International Conference Proceedings Series
- Full Text: false
- Reviewed:
- Description: Gene regulatory network (GRN) represents a set of genes and their regulatory interactions. The inference of the regulatory interactions between genes is usually carried out as an optimization problem using an appropriate mathematical model and the time-series gene expression data. Among the various models proposed for GRN inference, our recently proposed Michaelis-Menten kinetics based ODE model provides a good trade-off between the computational complexity and biological relevance. This model, like other known GRN models, also uses an evolutionary algorithm for parameter estimation. Since the search space for large networks is huge, leading to a low accuracy of inference, it is important to reduce the search region for improved performance of the optimization algorithm. In this paper, we propose a classification method using gene knockout data to eliminate a large infeasible region from the optimization search area. We also propose a method for partial inference of regulations when all the regulators of a given regulated gene are unregulated genes. The proposed method is evaluated by reconstructing in silico networks of large sizes. © 2018 ACM.
Passive detection of splicing and copy-move attacks in image forgery
- Authors: Islam, Mohammad , Kamruzzaman, Joarder , Karmakar, Gour , Murshed, Manzur , Kahandawa, Gayan
- Date: 2018
- Type: Text , Conference proceedings , Conference paper
- Relation: 25th International Conference on Neural Information Processing, ICONIP 2018; Siem Reap, Cambodia; 13th-16th December 2018; published in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) Vol. 11304 LNCS, p. 555-567
- Full Text:
- Reviewed:
- Description: Internet of Things (IoT) image sensors for surveillance and monitoring, digital cameras, smart phones and social media generate huge volume of digital images every day. Image splicing and copy-move attacks are the most common types of image forgery that can be done very easily using modern photo editing software. Recently, digital forensics has drawn much attention to detect such tampering on images. In this paper, we introduce a novel feature extraction technique, namely Sum of Relevant Inter-Cell Values (SRIV) using which we propose a passive (blind) image forgery detection method based on Discrete Cosine Transformation (DCT) and Local Binary Pattern (LBP). First, the input image is divided into non-overlapping blocks and 2D block DCT is applied to capture the changes of a tampered image in the frequency domain. Then LBP operator is applied to enhance the local changes among the neighbouring DCT coefficients, magnifying the changes in high frequency components resulting from splicing and copy-move attacks. The resulting LBP image is again divided into non-overlapping blocks. Finally, SRIV is applied on the LBP image blocks to extract features which are then fed into a Support Vector Machine (SVM) classifier to identify forged images from authentic ones. Extensive experiment on four well-known benchmark datasets of tampered images reveal the superiority of our method over recent state-of-the-art methods.
A rule based inference model to establish strategy-process relationship
- Authors: Dinh, Loan , Karmakar, Gour , Kamruzzaman, Joarder , Stranieri, Andrew
- Date: 2017
- Type: Text , Conference proceedings
- Relation: 30th International Business Information Management Association Conference - Vision 2020: Sustainable Economic development, Innovation Management, and Global Growth, IBIMA 2017; Madrid, Spain; 8th-9th November 2017 Vol. 2017-January, p. 4544-4556
- Full Text: false
- Reviewed:
- Description: An effective relationship between business processes and their relevant strategies helps enterprises achieve their goals. As a business organisation changes quickly, business processes implement their relevant business operations for efficiency. It is important to know which business process achieves which business strategies dynamically. To the best of our knowledge, there exists a framework which aims to automatically determine the strategy-process relationship (Morrison et al. 2011). However, this framework can only work when the effect of the business process is known, but it is difficult to determine such effect accurately. Moreover, by optimising business processes to satisfy business strategies, higher efficiency may be achieved but there is a high chance of losing discriminative information. It therefore creates certain level of uncertainty in achieving accurate strategy-process relationship. To reduce this uncertainty and determine the relationship accurately between business processes and their relevant strategies as defined by business domain experts, in this paper, we introduce a rule-based inference model. This model not only helps business organisations realize which business processes need to be involved for the organisation to achieve their goals when strategies are made, but also reduces the possibility of losing important details from business process optimisation. We have developed a business case to validate our proposed model and the results show that our model can infer the relation accurately for each rule defined for the related business case.
Dynamic content distribution for decentralized sharing in tourist spots using demand and supply
- Authors: Kamruzzaman, Joarder , Karmakar, Gour , Gondal, Iqbal , Kaisar, Shahriar
- Date: 2017
- Type: Text , Conference proceedings
- Relation: 13th IEEE International Wireless Communications and Mobile Computing Conference, IWCMC 2017; Valencia, Spain; 26th-30th June 2016 p. 2121-2126
- Full Text: false
- Reviewed:
- Description: Decentralized content sharing (DCS) is emerging as an important platform for sharing contents among smart mobile device users, where devices form an ad-hoc network and communicate opportunistically. Existing DCS approaches for tourist spot like scenarios achieve low delivery success rate and high latency as they do not focus on dynamic demand for contents which usually vary considerably with the number of visitors present or occurrence of some influencing events. The amount of available supply also changes because of the nodes leaving the area. Only way to improve content delivery service is to distribute the contents in strategic positions based on dynamic demand and supply. In this paper, we propose a dynamic content distribution (DCD) method considering dynamic demand and supply for contents in tourist spots. Simulation results validate the improvement of the proposed approach. © 2017 IEEE.
- Description: 2017 13th International Wireless Communications and Mobile Computing Conference, IWCMC 2017
Dynamically controlling exterior and interior window coverings through IoT for environmental friendly smart homes
- Authors: Karmakar, Gour , Roy, Soma , Chattopadhyay, Gopinath , Xiao, Zhigang
- Date: 2017
- Type: Text , Conference proceedings , Conference paper
- Relation: 2017 IEEE International Conference on Mechatronics, ICM 2017; Gippsland, Australia; 13th-15th February 2017 p. 487-491
- Full Text: false
- Reviewed:
- Description: Energy saving using smart home is of paramount importance to reduce heating and cooling energy consumption, and promote sustainable environment. Awnings and blinds have exhibited their effectiveness to reduce heating gain in summer and cooling loss in winter, respectively. Awnings are more effective to reduce heat gain in summer than blinds, while the opposite is true in winter. There exist many approaches in the current literature to remotely control flat curtains and blinds. However, up to our knowledge, no automatic technique is available in the literature, which can dynamically control the orientation of an exterior covering so that it can act like a blind in winter and an awning in summer. In this paper, we propose an automatic on-demand system to control the orientation and size of such exterior covering, and the turning air conditioners, heaters and lights on and off considering the rate of change of room temperature, and its lighting condition. We also discuss the properties and design of such exterior covering. A simulation model was developed to analysis the performance of our approach in terms of energy savings both in summer and winter. © 2017 IEEE.
- Description: Proceedings - 2017 IEEE International Conference on Mechatronics, ICM 2017
Exploiting evolving trust relationships in the modelling of opinion formation dynamics in online social networks
- Authors: Das, Rajkumar , Kamruzzaman, Joarder , Karmakar, Gour
- Date: 2017
- Type: Text , Conference proceedings , Conference paper
- Relation: 31st IEEE International Conference on Advanced Information Networking and Applications, AINA 2017; Taipei, Taiwan; 27th-29th March 2017 p. 872-879
- Full Text: false
- Reviewed:
- Description: Mass participation of the members of a society in discussions to resolve issues related to a topic leads to forming public opinion. The timeline of the underlying dynamics goes through several distinguishable phases, and experiences transition from one to another. After initiated by concerned individuals, it draws active attention from almost everyone, and with time progression, people's participation starts declining as the issues are resolved or lost attraction. The existing works in the literature to capture the opinion formation process pay attention to model the dynamics in its active phase and thus ignore the other phases and the corresponding phase transitions. Trust relationships among the participants dynamically shape their interactions in different stages of the dynamics. Existing works fail to incorporate trust in defining the extent of influence one has on others, as they define the social relationships in the opinion space. To address this issue, we adopt simulated annealing to model the transitional behaviour of the dynamics, and then, amalgamate peoples relationships in the trust space with that in the opinion space to define the meta-heuristics of the algorithm for capturing the dynamical properties of the process. Finally, through simulation, we observe that our model is insightful in representing peoples' evolving behaviour in the different stages of opinion formation process, and consequently, can capture the various properties of the steady-state outcomes of the dynamics. © 2017 IEEE.
- Description: Proceedings - International Conference on Advanced Information Networking and Applications, AINA
Significance level of a query for enterprise data
- Authors: Thi Ngoc Dinh, Loan , Karmakar, Gour , Kamruzzaman, Joarder , Stranieri, Andrew , Das, Rajkumar
- Date: 2017
- Type: Text , Conference proceedings
- Relation: 30th International Business Information Management Association Conference - Vision 2020: Sustainable Economic development, Innovation Management, and Global Growth, IBIMA 2017; Madrid, Spain; 8th-9th November 2017 Vol. 2017-January, p. 4494-4504
- Full Text: false
- Reviewed:
- Description: To operate enterprise activities, a large number of queries need to be processed every day through an enterprise system. Consequently, such a system frequently faces hugely overloaded information and incurs high delay in producing query responses for big data. This is because, traditional queries are normally treated with equal importance. With the advent of big data and its use in enterprise systems and the growth of process complexity, the traditional approach of query processing is no more suitable as it does not consider semantic information and captures all data irrespective of their relevance to a business organization, which eventually increases the computational time in both big data collection and analysis. The significance level of a query can make a trade-off between query response delay and the extent of data collection and analysis. This motivates us to concentrate on determining the significance level of a query considering its importance to an enterprise system. To our knowledge, no such approach is available in the literature. To bridge this research gap, this paper, for the first time, proposes an approach to determine the significance level of a query to prioritize them with the relevance to a business organization. As business processes play key roles in any enterprise system and all business processes are not equally important, this is done by determining the semantic similarity between a query and the processes of a business organization and the importance of a business process to that organization. With a case study on an enterprise system of a retail company, the results produced by our proposed approach have shown that significance level is higher for more important queries compared to the less important ones.
Carry me if you can : A utility based forwarding scheme for content sharing in tourist destinations
- Authors: Kaisar, Shahriar , Kamruzzaman, Joarder , Karmakar, Gour , Gondal, Iqbal
- Date: 2016
- Type: Text , Conference proceedings
- Relation: 22nd Asia-Pacific Conference on Communications, APCC 2016; Yogyakarta, Indonesia; 25th-27th August 2016 p. 261-267
- Full Text:
- Reviewed:
- Description: Message forwarding is an integral part of the decentralized content sharing process as the content delivery success highly depends on it. Existing literature employs spatio-temporal regularity of human movement pattern and pre-existing social relationship to take message forwarding decisions. However, such approaches are ineffectual in environments where those information are unavailable such as a tourist spot or camping site. In this study, we explore the message forwarding techniques in such environments considering the information that are readily available and can be gathered on the fly. We propose a utility based forwarding scheme to select the appropriate forwarder node based on co-location stay time, connectivity and available resources. A higher co-location stay time reflects that the forwarder and the destination node is likely to have more opportunistic contacts, while the connectivity and available resource ensure that the selected forwarder has sufficient neighbours and resources to carry the message forward. Simulation results suggest that the proposed approach attains high hit and success rate and low latency for successful content delivery, which is comparable to those proposed for work-place type scenarios with regular movement pattern and pre-existing relationships. © 2016 IEEE.
Exploiting temporal genetic correlations for enhancing regulatory network optimization
- Authors: Youseph, Ahammed , Chetty, Madhu , Karmakar, Gour
- Date: 2016
- Type: Text , Conference proceedings
- Relation: 23rd International Conference on Neural Information Processing, ICONIP 2016; Kyoto, Japan; 16th-21st October 2016; published in Neural Information Processing (Lecture Notes in Computer Science series) Vol. 9947 LNCS, p. 479-487
- Full Text: false
- Reviewed:
- Description: Inferring gene regulatory networks (GRN) from microarray gene expression data is a highly challenging problem in computational and systems biology. To make GRN reconstruction process more accurate and faster, in this paper, we develop a technique to identify the gene having maximum in-degree in the network using the temporal correlation of gene expression profiles. The in-degree of the identified gene is estimated applying evolutionary optimization algorithm on a decoupled S-system GRN model. The value of in-degree thus obtained is set as the maximum in-degree for inference of the regulations in other genes. The simulations are carried out on in silico networks of small and medium sizes. The results show that both the prediction accuracy in terms of well known performance metrics and the computational time of the optimization process have been improved when compared with the traditional S-system model based inference. © Springer International Publishing AG 2016.
- Description: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Who are convincing? An experience based opinion formation dynamics in online social networks
- Authors: Das, Rajkumar , Kamruzzaman, Joarder , Karmakar, Gour
- Date: 2016
- Type: Text , Conference proceedings , Conference paper
- Relation: 30th European Simulation and Modelling Conference, ESM 2016; Las Palmas, Spain; 26th-28th October 2016 p. 167-173
- Full Text: false
- Reviewed:
- Description: Online social network (OSN) is one of the major platforms where our opinions are formed now-a-days and increasing so. Opinion formation dynamics captures the ways public opinions are formed, mainly from two different sources, (i) neighbours' opinions, (ii) external opinions from sources other than the neighbours. In this paper, we formulate an opinion formation model by considering two very important factors, that were ignored or a very little explored in the literature. First, we model the convincing power of the opinions encountered from the two sources. Second, we incorporate the experience of users' previous interactions with the two opinion sources. The problem is formulated as an agent based model where each member of an OSN is represented with an agent and their relationships with a graph. Finally through simulation, we create various scenarios, and apply our model to observe the steady state outcomes of the dynamics. This helps us to study the nature of the public opinions under various influences of our model parameters.
- Description: European Simulation and Modelling Conference 2016, ESM 2016
Business context in big data analytics
- Authors: Dinh, Loan , Karmakar, Gour , Kamruzzaman, Joarder , Stranieri, Andrew
- Date: 2015
- Type: Text , Conference proceedings
- Relation: 10th International Conference on Information, Communications and Signal Processing, ICICS 2015; Singapore; 2nd-4th December 2015
- Full Text: false
- Reviewed:
- Description: Big data are generated from a variety of sources having different representation forms and formats, it raises a research question as how important data relevant to a business context can be captured and analyzed more accurately to represent deep and relevant business insight. There is a number of existing big data analytic methods available in the literature that consider contextual information such as the context of a query and its users, the context of a query-driven recommendation system, etc. However, these methods still have many challenges and none of them has considered the context of a business in either data collection or analysis process. To address this research gap, we introduce a big data analytic technique which embeds a business context in terms of the significance level of a query into the bedrock of its data collection and analysis process. We implemented our proposed model under the framework of Hadoop considering the context of a grocery shop. The results exhibit that our method substantially increases the amount of data collection and their deep insight with an increase of the significance level value. © 2015 IEEE.
- Description: 2015 10th International Conference on Information, Communications and Signal Processing, ICICS 2015
Consistency driven opinion formation modelling in presence of external sources
- Authors: Das, Rajkumar , Kamruzzaman, Joarder , Karmakar, Gour
- Date: 2015
- Type: Text , Conference proceedings
- Relation: International Joint Conference on Neural Networks, IJCNN 2015; Killarney, Ireland; 12th-17th July 2015
- Full Text: false
- Description: Opinion formation in social networks has changed in a more rigorous way due to the inception of Online Social Networks (OSNs) as a platform of generating and sharing huge amount of contents as well as easy and ubiquitous access to varied information sources. Our opinions are not only updated through interactions with our neighbours in OSNs, but also shaped by the opinions received from information sources external to the native OSNs. Current models only consider the neighbours' influence in opinion evolution, thus lack the impact of other information sources, e.g., news media, web search, bulletin board, discussion forum on opinion formation. They consider individual opinion distances to model the influence among interactive neighbours, but fail to capture the influence of majority supported opinions and its possible impact in opinion evolution. Our model explicitly captures the effect of external sources on opinion formation in an OSN. We combine the implication of most perceived opinions in terms of consistency along with opinion distance to emulate the influence of different opinion sources. Consistency is measured by the entropy of opinions derived from a particular source type. Simulation results show that our model properly captures the consensus, polarization and fragmentation properties of opinion evolution. Finally, we investigate the influence of stubborn agents on opinion formation and compare it with a contemporary model. © 2015 IEEE.