Search or add a thesis

Advanced Search (Beta)
Home > Softwaredefect Prediction Model Using Multi-Layer Feed Forwardneural Networks

Softwaredefect Prediction Model Using Multi-Layer Feed Forwardneural Networks

Thesis Info

Access Option

External Link

Author

Faseeha Matloob

Institute

Virtual University of Pakistan

Institute Type

Public

City

Lahore

Province

Punjab

Country

Pakistan

Thesis Completing Year

2019

Thesis Completion Status

Completed

Subject

Software Engineering

Language

English

Link

http://vspace.vu.edu.pk/detail.aspx?id=342

Added

2021-02-17 19:49:13

Modified

2024-03-24 20:25:49

ARI ID

1676721027895

Asian Research Index Whatsapp Chanel
Asian Research Index Whatsapp Chanel

Join our Whatsapp Channel to get regular updates.

Similar


Software defect prediction techniques are being focused by many researchers due to its effectiveness for cost reduction in testing process. Most of the software defect dataset contains uncleaned, noisy, high dimensional and imbalance data. These problems reduce the prediction accuracy of a classifier. In this paper, a framework is proposed which combines approaches to deal with all these problems. This framework comprises of four stages 1) data preprocessing 2) feature selection (FS) 3) class balancing, and 4) classification through ensemble learning. Normalization is performed on cleaned datasets. Multilayer perceptron (MLP) is used as subset evaluator in FS process with six search methods Best first (BF), Greedy Stepwise (GS), Genetic Search (GA), Particle swam Optimization Search (PSO), Rank Search (RS) and Linear forward selection (LFS). Resample and SMOTE algorithm are used for class balancing. In classification stacking ensemble model is applied to build model on 80% of the input data. Here meta classifier is set to MLP and base classifiers include decision tree (J48), Random forest (RF), Support vector machine (SVM), K nearest neighbor (kNN) and Bayes Net (BN). Parameter tuning of meta and base classifier is also performed. Performance is evaluated on NASA MDP using precision, recall, F-measure, MCC, ROC and accuracy. Results have shown that the proposed method showed significant improvement is defect prediction compared to base classifiers.
Loading...
Loading...

Similar Books

Loading...

Similar Chapters

Loading...

Similar News

Loading...

Similar Articles

Loading...

Similar Article Headings

Loading...

21. Al-Anbiya/The Prophets

21. Al-Anbiya/The Prophets

I/We begin by the Blessed Name of Allah

The Immensely Merciful to all, The Infinitely Compassionate to everyone.

21:01
The time of their reckoning draws ever closer to the people as it starts while experiencing their death,
and yet in their heedlessness of its coming, they keep turning away.

21:02
Whenever a new Message comes to them from their Rabb - The Lord,
they listen to it playfully.

21:03
Their hearts are preoccupied with trivial things.
Yet, concealing their inner thoughts, the unjust/wrongdoers say in their private conversations:
‘Is this Muhammad not a human being like you?
Will you, then, submit to his magic, with your eyes wide open?’

21:04
He - The Prophet – said:
‘My Rabb - The Lord Knows whatever you confer and conspire and, in fact, every word
spoken within the celestial realm and the terrestrial world.
For HE is the All-Listening, the All-Knowing.’

21:05
No way!
‘Instead, they allege:
The Qur’an is just a collection of his jumbled false dreams.
Perhaps he has fabricated it.
Rather he is a poet!’
If this is not the case, then let him bring us a miraculous Sign, as were the former Messengers assigned with Signs.

21:06
Not one of the habitation(s) which WE destroyed in punishment before they had believed.
So, will they, then, be willing to believe or get destroyed?

21:07
And WE have not assigned any Messenger before you except for those men on to whom WE revealed -
so just ask scholars of the Former Divine Scriptures, if you do not know.

21:08
And WE did not make for...

اختلاف امت سے متعلقہ احادیث کی روشنی میں راہ حق کی تعیین: ایک تحقیقی اور تنقیدی مطالعہ

A great deal of research work has been done upon the differences present in the Muslim Umm’ah and for its dissolution. Hence, scholars like Ashari, Baghdadi, Ibn al- Hazm and Shehristani has produced great critical and analytical studies, however the sect proclaiming on “Rah-e –Haq" (the truly guided path) needs critical review. In this article, a scholarly analysis and estimation of the "Rah-e-Haq"sect (the guided ones) is being attempted in the light of the relevant traditions.

Improving Fertilizer Use Efficiency and Productivity of Maize-Maize Crop Rotation by Organic Amendments

Intensive agriculture has degraded the soil and yield owing to reduction in organic matter. Increasing cost and nutrient losses, associated with the inorganic fertilizers always demand some alternate arrangements. Maize (being a C4 crop) based cropping systems exploit available soil nutrient resources in an exhaustive manner which ultimately reduces the fertility status of the soils. Integrated nutrient management approaches with green manuring (GM) and composted poultry manure (CPM) may improve this situation. Two years’ field experiments on maize, comprising four consecutive seasons (autumn 2016, spring 2017, autumn 2017 and spring 2018) were carried out at Agronomic Research area, University of Agriculture, Faisalabad, Pakistan. GM and CPM were used as tools to enhance soil productivity and agronomic nutrient efficiency in two experiments which were continued side by side. In first experiment, production potential of maize was evaluated under six levels of nitrogen i.e., 0%, 20%, 40%, 60%, 80% and 100% of recommended dose with and without GM. In second experiment, again performance of maize was observed under six levels of nitrogen (N), phosphorus (P) and potassium (K) i.e., 0, 20, 40, 60, 80 and 100% of their respective recommended dose in combination with two levels of CPM (control & 3 ton ha-1). Results showed that in both the experiments regarding N and NPK levels with GM and CPM respectively, maximum values of number of grain rows per cob, number of grains per cob, 1000-grain weight, grain yield and agronomic N efficiency were obtained where 100% rate was applied in autumn, 2016. However across the two years, statistically no difference were found in all plant growth results between 60% and 100% fertilizer rates, while the 60% rate was more economical than 100%. Application of N and NPK at 100% rate during both the experiments produced highest plant height, cob diameter and biological compared with lower levels. In the first experiment, GM improved grain yield in almost every season (21% in autumn 2016, 33% in spring 2017, 37% in autumn 2017 & 48% in spring 2018). Likewise, GM also resulted in substantial increase in soil organic matter (OM) (40% 67%, 88% & 88%), available N (55%, 140%, 210% & 219%), total P (14%, 69%, 112% & 128%) respectively. Similarly, the results of second experiment disclosed that CPM improved grain yield in all seasons (21% in autumn 2016, 26% in spring 2017, 36% in autumn 2017 & 51% in spring, 2018). CPM also resulted in significant increase in OM (30%, 84%, 41% & 120%), available N (60%, 166%, 219% & 231%), total P (35%, 58%, 89% & 127%), and total potassium (K) (14%, 15%, 23% & 26%) respectively. In conclusion, continuous application of GM/CPM for two years along with 60% N/NPK rates economized the cost of production, improved soil health and maize yield on sustainable basis.