Sadia Sharmin
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Sadia Sharmin has a total of 24 co-authors. Here are the top co-authors by number of shared publications:
Mohammad Shoyaib
University of Dhaka, Dhaka, Bangladesh
7 shared publicationsAmin Ahsan Ali
Independent University, Bangladesh, Dhaka, Bangladesh
3 shared publications- Amin Ahsan Ali
Center for Computational & Data Sciences (CCDS) Independent University Bangladesh
3 shared publications Muhammad Asif Hossain Khan
3 shared publications- Mohammad Shoyaib
University of Dhaka
2 shared publications Eusha Kadir
University of Dhaka, Dhaka, Bangladesh
1 shared publication
Latest Publications
- 5 May 2023
- Published by Springer Nature in International Journal of Information Security
- Vol. 22 (5) , 1355-1369
cFEM: a cluster based feature extraction method for network intrusion detection
- 5 May 2023
- Published by Springer Nature in International Journal of Information Security
- Vol. 22 (5) , 1355-1369
cFEM: a cluster based feature extraction method for network intrusion detection
- 6 May 2020
- Published by Springer Nature
- Vol. 12085, 71-83
A Proximity Weighted Evidential k Nearest Neighbor Classifier for Imbalanced Data
In k Nearest Neighbor (kNN) classifier, a query instance is classified based on the most frequent class of its nearest neighbors among the training instances. In imbalanced datasets, kNN becomes biased towards the majority instances of the training space. To solve this problem, we propose a method called Proximity weighted Evidential kNN classifier. In this method, each neighbor of a query instance is considered as a piece of evidence from which we calculate the probability of class label given feature values to provide more preference to the minority instances. This is then discounted by the proximity of the neighbor to prioritize the closer instances in the local neighborhood. These evidences are then combined using Dempster-Shafer theory of evidence. A rigorous experiment over 30 benchmark imbalanced datasets shows that our method performs better compared to 12 popular methods. In pairwise comparison of these 12 methods with our method, in the best case, our method wins in 29 datasets, and in the worst case it wins in least 19 datasets. More importantly, according to Friedman test the proposed method ranks higher than all other methods in terms of AUC at 5% level of significance.
- 6 May 2020
- Published by Springer Nature
- Vol. 12085, 71-83
A Proximity Weighted Evidential k Nearest Neighbor Classifier for Imbalanced Data
In k Nearest Neighbor (kNN) classifier, a query instance is classified based on the most frequent class of its nearest neighbors among the training instances. In imbalanced datasets, kNN becomes biased towards the majority instances of the training space. To solve this problem, we propose a method called Proximity weighted Evidential kNN classifier. In this method, each neighbor of a query instance is considered as a piece of evidence from which we calculate the probability of class label given feature values to provide more preference to the minority instances. This is then discounted by the proximity of the neighbor to prioritize the closer instances in the local neighborhood. These evidences are then combined using Dempster-Shafer theory of evidence. A rigorous experiment over 30 benchmark imbalanced datasets shows that our method performs better compared to 12 popular methods. In pairwise comparison of these 12 methods with our method, in the best case, our method wins in 29 datasets, and in the worst case it wins in least 19 datasets. More importantly, according to Friedman test the proposed method ranks higher than all other methods in terms of AUC at 5% level of significance.
- 6 May 2020
- Published by Springer Nature
- Vol. 12084, 830-842
Discretization and Feature Selection Based on Bias Corrected Mutual Information Considering High-Order Dependencies
Mutual Information (MI) based feature selection methods are popular due to their ability to capture the nonlinear relationship among variables. However, existing works rarely address the error (bias) that occurs due to the use of finite samples during the estimation of MI. To the best of our knowledge, none of the existing methods address the bias issue for the high-order interaction term which is essential for better approximation of joint MI. In this paper, we first calculate the amount of bias of this term. Moreover, to select features using \(\chi ^2\) based search, we also show that this term follows \(\chi ^2\) distribution. Based on these two theoretical results, we propose Discretization and feature Selection based on bias corrected Mutual information (DSbM). DSbM is extended by adding simultaneous forward selection and backward elimination (DSbM\(_\mathrm{fb}\)). We demonstrate the superiority of DSbM over four state-of-the-art methods in terms of accuracy and the number of selected features on twenty benchmark datasets. Experimental results also demonstrate that DSbM outperforms the existing methods in terms of accuracy, Pareto Optimality and Friedman test. We also observe that compared to DSbM, in some dataset DSbM\(_\mathrm{fb}\) selects fewer features and increases accuracy.- 6 May 2020
- Published by Springer Nature
- Vol. 12084, 830-842
Discretization and Feature Selection Based on Bias Corrected Mutual Information Considering High-Order Dependencies
Mutual Information (MI) based feature selection methods are popular due to their ability to capture the nonlinear relationship among variables. However, existing works rarely address the error (bias) that occurs due to the use of finite samples during the estimation of MI. To the best of our knowledge, none of the existing methods address the bias issue for the high-order interaction term which is essential for better approximation of joint MI. In this paper, we first calculate the amount of bias of this term. Moreover, to select features using \(\chi ^2\) based search, we also show that this term follows \(\chi ^2\) distribution. Based on these two theoretical results, we propose Discretization and feature Selection based on bias corrected Mutual information (DSbM). DSbM is extended by adding simultaneous forward selection and backward elimination (DSbM\(_\mathrm{fb}\)). We demonstrate the superiority of DSbM over four state-of-the-art methods in terms of accuracy and the number of selected features on twenty benchmark datasets. Experimental results also demonstrate that DSbM outperforms the existing methods in terms of accuracy, Pareto Optimality and Friedman test. We also observe that compared to DSbM, in some dataset DSbM\(_\mathrm{fb}\) selects fewer features and increases accuracy.- Open Access
- 9 September 2019
- Published by Association for Computing Machinery (ACM)
- p. 736-740
Can a simple approach identify complex nurse care activity?
- 9 September 2019
- Published by Association for Computing Machinery (ACM)
- p. 736-740
Open AccessCan a simple approach identify complex nurse care activity?
- 1 July 2019
- Published by Elsevier in Pattern Recognition
- Vol. 91, 162-174
Simultaneous feature selection and discretization based on mutual information
- 1 July 2019
- Published by Elsevier in Pattern Recognition
- Vol. 91, 162-174
Simultaneous feature selection and discretization based on mutual information
- 1 June 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 520-525
Bug Severity Classification Based on Class-Membership Information
Now-a-days bug classification along with its severity prediction has become an important issue for better maintenance and cost reduction of software. Different approaches have already been introduced in this regard including topic modeling, concept profile etc., however, most of them require a lot of parameter tuning, different types of computations for identifying bug severity. Furthermore, for reduced computational cost many of the state-of-the-art methods use standard Naïve Bayes for classification. However, priors used in this case may degrade the results due to the unavailability of training data. Moreover, cross project performance has rarely been discussed. To address these issues, we introduce a method namely Class-Membership Information of a Term (CMT) that does not use any priors, is computationally simple, free from parameter tuning and performs better compared to the existing methods. Rigorous experiments based on three benchmark datasets demonstrate that CMT on an average performs at most 5% and 12.5% better than other state-of-the-art methods considering within and cross project classification respectively.
- 1 June 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 520-525
Bug Severity Classification Based on Class-Membership Information
Now-a-days bug classification along with its severity prediction has become an important issue for better maintenance and cost reduction of software. Different approaches have already been introduced in this regard including topic modeling, concept profile etc., however, most of them require a lot of parameter tuning, different types of computations for identifying bug severity. Furthermore, for reduced computational cost many of the state-of-the-art methods use standard Naïve Bayes for classification. However, priors used in this case may degrade the results due to the unavailability of training data. Moreover, cross project performance has rarely been discussed. To address these issues, we introduce a method namely Class-Membership Information of a Term (CMT) that does not use any priors, is computationally simple, free from parameter tuning and performs better compared to the existing methods. Rigorous experiments based on three benchmark datasets demonstrate that CMT on an average performs at most 5% and 12.5% better than other state-of-the-art methods considering within and cross project classification respectively.
- 1 December 2017
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 750-754
BFSp: A feature selection method for bug severity classification
Bugs are inevitable in every software system. Resolving these bugs requires proper classification based on their severity. This is because the severe bugs need to be assigned to the developers immediately for fixing so that the vulnerability of software can be reduced. For this reason, several automated bug severity classification methods are already proposed to date that use the terms extracted from bug reports. However, most of the time they lack to identify the desired set of features that enhance the classification results. Besides, the existing works mainly focuses on classifying the bug severity within the different versions of a software. However, identifying the bugs in the newly released software is also crucial. To resolve these issues, we propose a method namely Bug Feature Selection method that mainly uses Pareto Optimality in order to find the most informative features. To evaluate the effectiveness of the proposed approach, we use the bug reports of three open source projects namely Eclipse, Mozilla and GCC. The experimental results show that our technique achieves promising performance compared to other state-of-the-art algorithms.
- 1 December 2017
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 750-754
BFSp: A feature selection method for bug severity classification
Bugs are inevitable in every software system. Resolving these bugs requires proper classification based on their severity. This is because the severe bugs need to be assigned to the developers immediately for fixing so that the vulnerability of software can be reduced. For this reason, several automated bug severity classification methods are already proposed to date that use the terms extracted from bug reports. However, most of the time they lack to identify the desired set of features that enhance the classification results. Besides, the existing works mainly focuses on classifying the bug severity within the different versions of a software. However, identifying the bugs in the newly released software is also crucial. To resolve these issues, we propose a method namely Bug Feature Selection method that mainly uses Pareto Optimality in order to find the most informative features. To evaluate the effectiveness of the proposed approach, we use the bug reports of three open source projects namely Eclipse, Mozilla and GCC. The experimental results show that our technique achieves promising performance compared to other state-of-the-art algorithms.
- 1 January 2017
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 20, 1-6
Feature Selection and Discretization based on Mutual Information
Feature selection and discretization have been considered to be an important research topic in the field of pattern recognition and data mining. However, addressing both these issues at a time is rarely discussed in the existing research. In this paper, these issues have been addressed by developing a heuristic namely discretization and selection of features based on mutual information (DSM). Experimental results on 15 datasets show that the proposed DSM outperforms a number of state-of-the-art feature selection or discretization algorithms. On average, its accuracy surpasses that of the best performing state-of-the-art algorithms by 5% using Support Vector Machine. Moreover, for datasets with a large number of features, it shows promising accuracies even with far less number of features than the other competing algorithms.
- 1 January 2017
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 20, 1-6
Feature Selection and Discretization based on Mutual Information
Feature selection and discretization have been considered to be an important research topic in the field of pattern recognition and data mining. However, addressing both these issues at a time is rarely discussed in the existing research. In this paper, these issues have been addressed by developing a heuristic namely discretization and selection of features based on mutual information (DSM). Experimental results on 15 datasets show that the proposed DSM outperforms a number of state-of-the-art feature selection or discretization algorithms. On average, its accuracy surpasses that of the best performing state-of-the-art algorithms by 5% using Support Vector Machine. Moreover, for datasets with a large number of features, it shows promising accuracies even with far less number of features than the other competing algorithms.
- Open Access
- 22 March 2016
- Published by Association for Computing Machinery (ACM)
- p. 1-6
Software Defect Prediction Using Feature Space Transformation
In software quality estimation research, software defect prediction is a key topic. A defect prediction model is generally constructed using a variety of software attributes and each attribute may have positive, negative or neutral effect on a specific model. Selection of an optimal set of attributes for model development remains a vital yet unexplored issue. In this paper, we have introduced a new feature space transformation process with a normalization technique to improve the defect prediction accuracy. We proposed a feature space transformation technique and classify the instances using Support Vector Machine (SVM) with its histogram intersection kernel. The proposed method is evaluated using the data sets from NASA metric data repository and its application demonstrates acceptable accuracy.
- 22 March 2016
- Published by Association for Computing Machinery (ACM)
- p. 1-6
Open AccessSoftware Defect Prediction Using Feature Space Transformation
In software quality estimation research, software defect prediction is a key topic. A defect prediction model is generally constructed using a variety of software attributes and each attribute may have positive, negative or neutral effect on a specific model. Selection of an optimal set of attributes for model development remains a vital yet unexplored issue. In this paper, we have introduced a new feature space transformation process with a normalization technique to improve the defect prediction accuracy. We proposed a feature space transformation technique and classify the instances using Support Vector Machine (SVM) with its histogram intersection kernel. The proposed method is evaluated using the data sets from NASA metric data repository and its application demonstrates acceptable accuracy.
- 1 December 2015
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 184-189
SAL: An effective method for software defect prediction
For software quality assurance, software defect prediction (SDP) has drawn a great deal of attention in recent years. Its goal is to reduce verification cost, time and effort by predicting the defective modules efficiently. In SDP, proper attribute selection plays a significant role. However, selection of proper attributes and their representation in an efficient way are very challenging due to the lacking of standard set of attributes. To address these issues, we introduce Selection of Attribute with Log filtering (SAL) to select a proper set of attributes. Our proposed attribute selection process can effectively select the best set of attributes, which are relevant for the discrimination of defected and non-defected software modules. Further, we adopt log filtering to pre-process the input data. We have evaluated the proposed attribute selection method using several widely used publicly available datasets. The simulation results demonstrate that our method is more effective to improve the accuracy of SDP than the existing state-of-the-art methods.
- 1 December 2015
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 184-189
SAL: An effective method for software defect prediction
For software quality assurance, software defect prediction (SDP) has drawn a great deal of attention in recent years. Its goal is to reduce verification cost, time and effort by predicting the defective modules efficiently. In SDP, proper attribute selection plays a significant role. However, selection of proper attributes and their representation in an efficient way are very challenging due to the lacking of standard set of attributes. To address these issues, we introduce Selection of Attribute with Log filtering (SAL) to select a proper set of attributes. Our proposed attribute selection process can effectively select the best set of attributes, which are relevant for the discrimination of defected and non-defected software modules. Further, we adopt log filtering to pre-process the input data. We have evaluated the proposed attribute selection method using several widely used publicly available datasets. The simulation results demonstrate that our method is more effective to improve the accuracy of SDP than the existing state-of-the-art methods.