Ghulam Ishaq Khan Institute of Engineering Sciences and Technology

About: Ghulam Ishaq Khan Institute of Engineering Sciences and Technology is a education organization based out in Topi, Pakistan. It is known for research contribution in the topics: Thin film & Quantum efficiency. The organization has 618 authors who have published 940 publications receiving 10674 citations.

Employing artificial neural networks for constructing metadata-based model to automatically select an appropriate data visualization technique

Abstract: Display Omitted Solution to automatically select appropriate visualization technique based on metadata is presented.A purpose built dataset extracted from existing knowledge in the field is used to train classifiers.A comparison of the results obtained from the best ANN architecture is performed with five other classifiers.The proposed system outperforms four classifiers in terms of accuracy and five classifiers based on running time.The work brings new perspective in the field of visualization. Advances in computing technology have been instrumental in creating an assortment of powerful information visualization techniques. However, the selection of a suitable and effective visualization technique for a specific dataset and a data mining task is not trivial. This work automatically selects an appropriate visualization technique based on the given metadata and the task that a user intends to perform. The appropriate visualization is predicted based on an artificial neural network (ANN)-based model which classifies the input data into one of the eight predefined classes. A purpose built dataset extracted from the existing knowledge in the discipline is utilized to train the neural network. The dataset covers eight visualization techniques, including: histogram, line chart, pie chart, scatter plot, parallel coordinates, map, treemap, and linked graph. Various architectures using different numbers of hidden units, hidden layers, and input and output data formats have been evaluated to find the optimal neural network architecture. The performance of neural networks is measured using: confusion matrix, accuracy, precision, and sensitivity of the classification. Optimal neural network architecture is determined by convergence time and number of iterations. The results obtained from the best ANN architecture are compared with five other classifiers, k-nearest neighbor, nave Bayes, decision tree, random forest, and support vector machine. The proposed system outperforms four classifiers in terms of accuracy and all five classifiers based on execution time. The trained neural network is also tested on twenty real-world benchmark datasets, where the proposed approach also provides two alternate visualizations, in addition to the most suitable one, for a particular dataset. A qualitative comparison with the state-of-the-art approaches is also presented. The results show that the proposed technique assists in selecting an appropriate visualization technique for a given dataset with high accuracy.

A novel robust kernel for classifying high-dimensional data using Support Vector Machines

Abstract: This paper presents a new semantic kernel for classification of high-dimensional data in the framework of Support Vector Machines (SVM). SVMs have gained widespread application due to their relatively higher accuracy. The efficacy of SVMs, however, depends upon the separation of the data itself as well as the kernel function. Text data, for instance, is difficult to classify due to synonymy and polysemy in its contents, having multi-topical instances that can result in mislabeling, and being highly sparse in the bag-of-words representation. While the soft margin parameter and kernel tricks are used in SVM to deal with outliers and non-linearly separable data, using data statistics and correlation has not been fully explored in the literature. This paper explore the use co-similarity (i.e., soft co-clustering) to find latent relationships between documents motivated by the success of co-clustering and subspace clustering methods. It has been shown that the use of weighted higher-order paths between instances in the data can be a good measure of similarity values which can then be used for both classification and to correct mislabeled (or outlier) data in the training set. The proposed kernel is generic in nature and suitable for sparse, dyadic data where direct co-occurrences are not necessary common as in the case of textual data, link-analysis in social media networks, co-authorship, etc. It also studies the impact of noise in the training data and provides a technique to re-label such instances. It is also observed that re-labelling of selected training data reduces the adverse effect of outliers or label noise and can greatly improve the classification of the test data. To the best of our knowledge, we are the first to introduce a supervised co-similarity based kernel function and also provide mathematical formulation to show that it is a valid Mercer's kernel. Our experiments show that the proposed framework outperforms current and state-of-the-art methods in terms of classification accuracy and is more resilient to label noise.

Sustainable hybrid electric vehicle selection in the context of a developing country

Abstract: The ever increasing global warming is affecting both the environment and quality of life. The dependency on the usage of fossil fuels for transportation and power generation sector is harming the environment in terms of greenhouse gas (GHG) emissions. To limit the use of fossil fuel, the world has to move towards a renewable, clean, and economical form of energy. In the transportation sector, the paradigm shift towards electric mobility is a step towards the same goal. For a developing country like Pakistan, due to the lacking charging infrastructure, load shedding of electricity, and high cost of non-renewable electrical energy, a country like Pakistan cannot go dependent on fully electric vehicles (EVs). The country has to shift from normal internal combustion engine vehicles (ICEVs) to hybrid electric vehicles (HEVs). This paper aims to select the most sustainable HEV in the context of a developing country, Pakistan. Using the multi-criteria decision making (MCDM) technique, i.e., fuzzy Technique for Order Performance by Similarity to Ideal Solution (TOPSIS), based on ten criteria and seven alternatives, it has been found that Toyota Aqua outperforms among all the other alternatives in terms of economic, social, and environmental perspective. Furthermore, to move towards hybrid technology, the government has to give relaxation in terms of customs duty and should encourage auto manufacturers to set up local industries of such vehicles in the country.

Measurement of Ultralow Level Bioethanol Concentration for Production Using Evanescent Wave Based Optical Fiber Sensor

Abstract: We report a U-bend plastic optical fiber (POF) sensor for measuring ultralow concentration of ethanol corresponding to the initial bioethanol production rate by cyanobacteria, i.e., 0.1–0.5 gL $^{\mathrm {\mathbf {-1}}}$ day $^{\mathrm {\mathbf {-1}}}$ . This production rate corresponds to 0.00499–0.0499 %wt of ethanol in total solution in terms of weight percent concentration. Refractive indices for these minute ethanol concentration values are not available in the literature hence mathematical estimation of the refractive indices for the ethanol-water solutions in this concentration range is demonstrated. The sensing principle is based on optical fiber-based evanescent wave absorption. Parameters affecting the response of an evanescent wave absorption sensor are analyzed for the intended concentration range of ethanol. Experimental results using the U-bend evanescent wave POF sensor are also presented for ethanol-water solutions having refractive indices corresponding to the bioethanol production rate. The excellent repeatability of the measurement is established and these real-time measurements show that sensor has a sensitivity of 817.76 O.D/RIU (O.D refers to optical density, unit of absorbance) with a 99.76% linearity. The limit of detection of the sensor is $9.2\times 10^{\mathrm {\mathbf {-7}}}$ RIU. It is also proved using refractive index calculations of ethanol-water solutions that the sensor exhibits a resolution of $10^{\mathrm {\mathbf {-7}}}$ RIU. The sensor of this investigation, therefore, represents a potential solution for online and real-time monitoring of the production of biofuels even at the very low-level initial concentration of product.

Controller design using fuzzy logic for a twin rotor MIMO system

Abstract: The paper discusses the design of a fuzzy logic controller for a twin rotor MIMO system. The system has three degrees of freedom, but we have designed the controller for two degrees of freedom (DOF) only, i.e., travel and elevation. The controller is designed such that a change in one degree should have minimum effect on the other and overall the system should remain stable. We have designed two fuzzy logic controllers: one for travel control and one for elevation: both are designed in MATLAB and have Sugeno inference. For the twin rotor MIMO system, commonly known as the helicopter problem, many conventional controllers exist. Our controller's performance had to be better than, or at least equal to, their's. To test the controller on a practical system was not possible, therefore we used simulations to benchmark the performance of the fuzzy logic controller. The simulation results show that the system performance with a fuzzy logic controller is superior to the system performance with a conventional PID or LQR controller, over the whole range of operating conditions.