Search or add a thesis

Advanced Search (Beta)
Home > Feature Selection Using Rough Set Based Heuristic Dependency Calculation

Feature Selection Using Rough Set Based Heuristic Dependency Calculation

Thesis Info

Access Option

External Link

Author

Raza, Muhammad Summair

Program

PhD

Institute

National University of Sciences & Technology

City

Islamabad

Province

Islamabad

Country

Pakistan

Thesis Completing Year

2018

Thesis Completion Status

Completed

Subject

Software Engineering

Language

English

Link

http://prr.hec.gov.pk/jspui/bitstream/123456789/10443/1/M%20Summair%20Raza_SE_2018_NUST_PRR.pdf

Added

2021-02-17 19:49:13

Modified

2024-03-24 20:25:49

ARI ID

1676727758249

Similar


The amount of data to be processed is significantly increasing day by day. The increase in data size is not only due to more number of records but also due to substantial number of attributes added to space. The phenomenon is leading to the dilemma called curse of dimensionality i.e. datasets with exponential number of attributes. The ideal approach is to reduce the number of dimensions such that resulted reduced set contains the same information as present in the entire set of attributes. There are various approaches to perform this task of dimensionality reduction. Recently, rough set-based approaches, which use attribute dependency to carry out feature selection, have been prominent. However, this dependency measure requires the calculation of the positive region, which is a computationally expensive task. In this research, we have proposed a new concept called the “Dependency Classes”, which calculates the attribute dependency without using the positive region. Dependency classes define the change in attribute dependency as we move from one record to another. By avoiding the positive region, they can be an ideal replacement for the conventional dependency measure in feature selection algorithms, especially for large datasets. A comparison framework was devised to measure the efficiency and effectiveness of the proposed measure. Experiments on various publically available datasets show that the proposed approaches provide significant computational performance with same accuracy as provided by conventional approach. We have also recommended seven feature selection algorithms using this measure. The experimental results have shown that algorithms using the classes were more effective than their counterparts using the positive region-based approach in terms of accuracy, execution time and required runtime memory.
Loading...
Loading...

Similar Books

Loading...

Similar Chapters

Loading...

Similar News

Loading...

Similar Articles

Loading...

Similar Article Headings

Loading...

نکی جہی ہک گل ہے یارو

نکی جہی ہک گل ہے یارو
عشق رونا ہر پل ہے یارو

حسن دی نگری دے وچہ عاشق
بیٹھا مار پتھل ہے یارو

چوٹ عشق دی جد آلگی
رون دا آیا ول ہے یارو

ہِکو یار سجن دا ملنا
اس مسئلے دا حل ہے یارو

یاد اوہدی پل پل انگاری
بھانبڑ جاندا بل ہے یارو

کالی زلف تے اکھ مستانی
بدن سارا مخمل ہے یارو

میں تے اس نوں چھڈنا ناہیں
میرا قول اٹل ہے یارو

دکھاں درداں ہاڑیاں والا
ہکو حل وصل ہے یارو

کوئی منصور سولی تے چڑھیا
شمس لہاندا کھل ہے یارو

سوہنی وچ چنابے ڈبی
سسی مارو تھل ہے یارو

جو کچھ کرنا ای اجّو کر لے
ڈٹھی کس نے کل ہے یارو

سیدحسن غزنویؒ کی دینی و علمی خدمات کا جائزہ

Syed Hassan of Ghaznain was an ancient Persian poet from 1078. We knew very little about his life and poetry until Dr. Ghulam Mustafa Khan (1912-2005), a renowned scholar did his extensive Ph.D. Thesis on him from Nagpur University in 1946. During his research, he visited several libraries of the subcontinent as well as brought together Hassan's poetry collection from London and Paris. Hassan Ghaznavi was a court poet of Bahram Shah Ghaznavi and also spent some time which Sultan Sanjar of Khorasan. The references of his life and beautiful poetry are mentioned in this article.

Miniaturized Reconfigurable Dielectric Resonator Antenna

The tremendous growth of mobile wireless devices in various form factors requires efficient, compact, and multi-function antennas, capable of providing high data rates at multiple application bands. Modern wireless devices incorporate a number of different antennas to provide services for cellular voice, video, data, Wi-Fi, and GPS connectivity, across multiple bands. High data rate requirements also demand the design of multiple input multiple output antennas in order to implement spatial multiplexing and diversity schemes. Reconfigurable antennas possess the ability to change their characteristics (frequency, pattern, polarization) as per application demand. Dynamic alteration of the radiation characteristics makes them a suitable candidate for multi-function, small size wireless devices, where the requirement of multiple antennas can be circumvented by a single reconfigurable antenna. Dielectric resonator antennas (DRAs) offer a number of benefits over the conventional conductor-based antennas. The advantages include high radiation efficiency, small size, and a versatility of feed mechanisms. Despite many benefits, however, the size of the DRA must be excessively miniaturized to make their integration feasible in wireless devices of small size and various form factors. The aim of this work is to devise some novel reconfiguration and simultaneous miniaturization techniques for DRA based antennas. Hybrid structures comprising of DR and other resonant structures (like microstrip patch, slots) have been explored previously for enhancement of impedance bandwidth, multi-band, and wideband applications. Hybrid antenna structures combine the desirable features of both resonators and are much beneficial to meet the diverse application demands. This work proposes a hybrid combination of DR element, with ground plane (GP) slots to design compact, efficient, and multi-function reconfigurable antennas. The reconfiguration properties of the proposed antenna structures satisfy multiple application band requirements, while the miniaturization aspects assist to save valuable space in small devices. The GP slots can either be employed as a resonant structure or act as a defected ground structure (DGS) for the resonant DR element. The proposed technique has been applied for the design of frequency reconfigurable (FR), single-input-single-output (SISO), multiple-input-multiple-output (MIMO), and frequency plus pattern reconfigurable antenna designs. A few novel designs are presented in the current dissertation that effectively combine the GP slots and the DR resonances to obtain efficient, multi-band, antenna structures for SISO as well as MIMO applications. The combination of probe-fed and microstrip line fed DR elements with GP slots has thoroughly been analyzed and designed with an emphasis to get higher tuning ranges and better isolation. In almost all designs, the reconfiguration aspect has been incorporated by placing switches on the GP slots, which in turn alter the slot length and result in achieving either frequency or pattern reconfiguration.