Search or add a thesis

Advanced Search (Beta)
Home > Optimization of Classifiers Using Genetic Programming

Optimization of Classifiers Using Genetic Programming

Thesis Info

Access Option

External Link

Author

Majid, Abdul

Program

PhD

Institute

Ghulam Ishaq Khan Institute of Engineering Sciences and Technology

City

Swabi

Province

KPK

Country

Pakistan

Thesis Completing Year

2006

Thesis Completion Status

Completed

Subject

Computer Science

Language

English

Link

http://prr.hec.gov.pk/jspui/handle/123456789/315

Added

2021-02-17 19:49:13

Modified

2024-03-24 20:25:49

ARI ID

1676727804131

Asian Research Index Whatsapp Chanel
Asian Research Index Whatsapp Chanel

Join our Whatsapp Channel to get regular updates.

Similar


The success of pattern classification system depends on the improvement of its classification stage. The work of thesis has investigated the potential of Genetic Programming (GP) search space to optimize the performance of various classification models. In this thesis, two GP approaches are proposed. In the first approach, GP is used to optimize the performance of individual classifiers. The performance of linear classifiers and nearest neighbor classifiers is improved during GP evolution to develop a high performance numeric classifier. In second approach, component classifiers are trained on the input data and their predictions are extracted. GP search space is then used to combine the predictions of component classifiers to develop an optimal composite classifier (OCC). This composite classifier extracts useful information from its component classifiers during evolution process. In this way, the decision space of composite classifier is more informative and discriminant. Effectiveness of GP combination technique is investigated for four different types of classification models including linear classifiers, support vector machines (SVMs) classifiers, statistical classifiers and instance based nearest neighbor classifiers. The successfulness of such composite classifiers is demonstrated by performing various experiments, while using Receiver Operating Characteristics (ROC) curve as the performance measure. It is evident from the experimental results that OCC outperforms its component classifiers. It attains high margin of improvement at small feature sets. Further, it is concluded that classification models developed by heterogeneous combination of classifiers have more promising results than their homogenous combination. GP optimization technique automatically caters the selection of suitable component classifiers and model selection. Two main objectives are achieved, while using GP optimization. First, objective achieved is the development of more optimal classification models. The second one is the enhancement in the GP search strategy itself.
Loading...
Loading...

Similar Books

Loading...

Similar Chapters

Loading...

Similar News

Loading...

Similar Articles

Loading...

Similar Article Headings

Loading...

ویکھو کیڈا ہے یار کمال کردا

ویکھو کیڈا ہے یار کمال کردا
آپے وچھڑدا، آپ ملال کردا
گیوں گیسو دراز نیں موہڈیاں تے
کوئی جا کے اوہنوں سوال کردا
اینویں پھردا جنگلاں وچ کملا
پہلے اندروں پرت کے بھال کردا
اینویں دوس ہے بندیاں ماڑیاں تے
چاہندا جو سو جاہ و جلال کردا
پہلے سچ تے جھوٹ نتار لیندا
جو چاہندا سو ساڈے نال کردا
جے کر اپنے مار نہ سٹ جاندے
غیر کردا، کی اوہدی مجال کردا
جند جان تے مال سب وار دیندے
اک وار جو خط ارسال کردا
منہ ویکھدا رتا شریک دا جے
نال چاٹاں دے اپنا لال کردا

پاکستانی دستور میں حکمرانوں کا تصور استحقاق اسلامی تعلیمات کی روشنی میں

Constitution is the basic code of every state system. There are laws for state administration, discipline and rulers in constitution. There are some privileges for the rulers in the Pakistani constitution. Among the privileges that Pakistani rulers have, laws of exception, protocol and luxury packages or facilities are included. In Pakistani constitution, the rulers also enjoy these privileges and according to the rules and regulation of parliament and senate. Our Constitution does not provide Parliamentarians any specific immunity against criminal actions as has been granted to the President as well as the Prime Minster. The only specific protection enjoyed by a Minister/Prime Minister is for official actions under powers of their office. The right of lessen or amendment in Sharīʻah penalty of the President of Pakistan is not right according to the Islamic Sharīʻah. But he can utilize honorary rights in criminological penalty. The Governor has the authority to dissolve the Provincial Assembly under certain circumstances during the emergency situation. If the ruler uses the option of freedom of opinion with deception, dishonesty and contempt of court then he should also be answerable. Sometimes rulers misuse their privileges and even exceed their powers. Discretionary options of the rulers must be under public interests. Such privilege rules must be amended which reflect inequality between rulers and masses. It is necessary to put the honorary rights of rulers under logic and there must be a law of behold for the unlawful usage of authority, so that the bad utilization of these laws can be prevented. Such reserved rights must be amended which enhance the concept of un-equity between the rulers and public. For the better administration harmony among the public and administration is necessary. In this research paper we will analyze the concept and importance of privileges mentioned in the Constitution of Pakistan in the light of Islamic teachings.

Fortification of Olive Oil in Palm Oil to Upgrade its Quality, Stability and Acceptibility

Oil Blending has been expected as mainly important solutions in the production of vegetable oil with good quality storage Constance’s and most favorable fatty acid composition. The current study was carried out to identify the finest oil blend in terms of physicochemical properties between palm and olive oils. Laboratory experiment on the “Fortification of olive oil in palm oil to upgrade its quality, stability and acceptability” was carried out at Nuclear Institute for Food and Agriculture Tarnab, Peshawar and also at The University of Agriculture Peshawar KP-Pakistan during 2014-15. Olive/palm oils were treated in sole or admixture at various ratios. Samples were tested with potato food frying at monthly interval continued for three months. Data were recorded on Peroxide value (PV), Free fatty acids (FFA), Anisidine value (AV), Iodine value (IV), Color index (CI), Beta carotene (βC) and Fatty acid profile of the oils. The results showed considerable variations in all the studied parameters tested in different blends under various stress conditions. Maximum PV (10.76 meq/kg), FFA (2.44%), AV (2.29), IV (77.09 g/100g) and CI (0.609) were recorded in sole olive oil, while minimum PV (9.69 meq/kg), FFA (0.65%), AV (0.88), IV (53.40 g/100g) and CI (0.146) were noted in sole palm oil. The samples kept in sunlight showed maximum PV (12.82 meq/kg), FFA (1.32%) and AV (1.39), while maximum IV (63.10 g/100g), CI (0.371) and BC (11.30) value were observed in samples stored at ambient temperature. Storage conditions increased PV from 2.18 to 27.82 meq/kg, FFA from 0.88 to 1.47%, AV from 1.02 to 1.56, while decreased IV from 63.38 to 58.05 g/100g, CI from 0.392 to 0.337 and BC from 27.87 to 3.46.After the frying process, maximum PV (8.37 meq/kg), FFA (2.24%), AV (2.28) and IV (71.69 g/100g) were noted in sole olive oil, while minimum PV (7.05 meq/kg), FFA (1.32%), AV (1.28) and IV (51.01 g/100g) were recorded in sole palm oil. Maximum PV (12.44 meq/kg), FFA (2.12%), AV (2.10), CI (0.392), while minimum IV (53.11 g/100g) and BC (23.14) were recorded after 3rd frying. As saturated fatty acid, palmitic acid was higher as 4.70, 31.80, 26.77, 24.85, 21.77 and 19.94in OO, PO, 10:90%, 20:80%, 40:60% and 60:40% oil blends, respectively. While, Eicosadienoic acid was higher as 0.912, 0.905, 0.895, 0.889 and 0.880in PO, 10:90%, 20:80%, 40:60% and 60:40% oil blends, respectively. Oleic acid was higher as 44.897, 31.096, 26.17, 23.756, 18.020 and 18.885 in OO, PO, 10:90%, 20:80%, 40:60% and 60:40% oil blends, respectively. Saturated fatty acids decreased from 0.781 to 0.494, from i i 2.889 to 2.834, from 2.586 to 2.217, from 2.376 to 2.092, from 2.173 to 1.797, from 1.940 to 1.598 in OO, PO, 10:90%, 20:80%, 40:60% and 60:40% oil blends, respectively after 3rd frying. Unsaturated fatty acids decreased from 1.039 to 0.854, from 0.209 to 0.143, from 0.204 to 0.140, from 0.200 to 0.136, from 0.195 to 0.132, from 0.179 to 0.120 in OO, PO, 10:90%, 20:80%, 40:60% and 60:40% oil blends, respectively after 3rd frying. Cis fatty acids decreased from 15.77 to 8.47, from 8.73 to 7.10, from 7.35 to 5.97, from 6.69 to 5.63, from 6.88 to 2.76, from 5.47 to 4.49 in OO, PO, 10:90%, 20:80%, 40:60% and 60:40% oil blends, respectively after 3rd frying. Maximum appearance (6.90), taste (7.57), texture (8.00), color (7.67) and overall acceptability (7.56) were noted in 10:90% oil blend. It was observed that low temperature storage was beneficial for oil quality in addition to blending of OO:PO at 10:90% and 20:80% ratios showed low PV and FFA value. Fatty acid profile showed that it has high monounsaturated fatty acids conflicting to oxidation. It is concluded that OO:PO oils mixed at 10:90% and 20:80% ratios is more economical and acceptable for improving the quality and stability of edible oils.