Analogy-based software project effort estimation. Contributions to projects similarity measurement, attribute selection and attribute weighting algorithms for analogy-based effort estimation.
Publication date
2010-10-01T15:31:18ZSupervisor
Neagu, DanielCowling, Peter I.
Keyword
Analogy-based effort estimationAttributes relevancy
Fuzzy Grey Relational Analysis
Fuzzy numbers
Attribute weighting
Attribute selection
Algorithms
Rights
The University of Bradford theses are licenced under a Creative Commons Licence.
Institution
University of BradfordDepartment
Department of Computing School of Computing, Informatics & MediaAwarded
2010Metadata
Show full item recordAbstract
Software effort estimation by analogy is a viable alternative method to other estimation techniques, and in many cases, researchers found it outperformed other estimation methods in terms of accuracy and practitioners¿ acceptance. However, the overall performance of analogy based estimation depends on two major factors: similarity measure and attribute selection & weighting. Current similarity measures such as nearest neighborhood techniques have been criticized that have some inadequacies related to attributes relevancy, noise and uncertainty in addition to the problem of using categorical attributes. This research focuses on improving the efficiency and flexibility of analogy-based estimation to overcome the abovementioned inadequacies. Particularly, this thesis proposes two new approaches to model and handle uncertainty in similarity measurement method and most importantly to reflect the structure of dataset on similarity measurement using Fuzzy modeling based Fuzzy C-means algorithm. The first proposed approach called Fuzzy Grey Relational Analysis method employs combined techniques of Fuzzy set theory and Grey Relational Analysis to improve local and global similarity measure and tolerate imprecision associated with using different data types (Continuous and Categorical). The second proposed approach presents the use of Fuzzy numbers and its concepts to develop a practical yet efficient approach to support analogy-based systems especially at early phase of software development. Specifically, we propose a new similarity measure and adaptation technique based on Fuzzy numbers. We also propose a new attribute subset selection algorithm and attribute weighting technique based on the hypothesis of analogy-based estimation that assumes projects that are similar in terms of attribute value are also similar in terms of effort values, using row-wise Kendall rank correlation between similarity matrix based project effort values and similarity matrix based project attribute values. A literature review of related software engineering studies revealed that the existing attribute selection techniques (such as brute-force, heuristic algorithms) are restricted to the choice of performance indicators such as (Mean of Magnitude Relative Error and Prediction Performance Indicator) and computationally far more intensive. The proposed algorithms provide sound statistical basis and justification for their procedures. The performance figures of the proposed approaches have been evaluated using real industrial datasets. Results and conclusions from a series of comparative studies with conventional estimation by analogy approach using the available datasets are presented. The studies were also carried out to statistically investigate the significant differences between predictions generated by our approaches and those generated by the most popular techniques such as: conventional analogy estimation, neural network and stepwise regression. The results and conclusions indicate that the two proposed approaches have potential to deliver comparable, if not better, accuracy than the compared techniques. The results also found that Grey Relational Analysis tolerates the uncertainty associated with using different data types. As well as the original contributions within the thesis, a number of directions for further research are presented. Most chapters in this thesis have been disseminated in international journals and highly refereed conference proceedings.Type
ThesisQualification name
PhDCollections
Related items
Showing items related by title, author, creator and subject.
-
Impact of material attributes & process parameters on critical quality attributes of the amorphous solid dispersion products obtained using hot melt extrusionThe feasibility of hot melt extrusion (HME) was explored for development of amorphous solid dispersion systems. Controlled release formulations were developed using a cellulose based derivative, AffinisolTMHPMC 100cP and 4M grades. BCS class II drugs ibuprofen and posaconazole were selected due to their difference in glass transition temperature and lipophilicity. This study focused on investigation of the impact the material attributes and process parameters on the critical quality attributes in preparation of amorphous solid dispersions using hot melt extrusion. The critical quality attributes were sub divided into three main attributes of material, process and product. Rheology of ibuprofen-Affinisol 100cP from melt phase to extrudate phase was tracked. A partial factorial design was carried out to investigate the critical parameters affecting HME. For optimisation of 40%IBU-Affinisol 100cP blends, a feed rate of 0.6kg/hr, screw speed of 500rpm and screw configuration with two mixing elements were found to be optimum for single phase extrudates. ATR-FTIR spectroscopy was found to be an indirect technique of choice in predicting the maximum ibuprofen drug load within extrudates. Prediction was based on the prepared extrudates without charging them to stability conditions. An alternative strategy of incorporation of di-carboxylic acids to increase the dissolution of posaconazole-Affinisol 4M blends was investigated. Succinic acid and L- malic acid incorporation was found to increase the dissolution of posaconazole. Although, the extrudates crystallised out quicker than the naïve posaconazole-Affinisol 4M, but free posaconazole formed eutectic and co-crystal with succinic and L-malic acid within extrudates. This lead to an increase in dissolution of the extrudates compared to day 0.
-
Evaluation of the Effect of Critical Process and Formulation Parameters on the Attributes of Nanoparticles Produced by Microfluidics. Design of Experiments Approach for Optimisation of Process and Formulation Parameters Affecting the Fabrication of Nanocrystals of Poorly Water-Soluble Drug Using Anti-solvent Precipitation in MicrofluidicAdvanced drug delivery systems have shown immense success through nanotechnology which overcomes the challenges posed by large sized particles such as poor solubility, bioavailability, absorption, and target-specific delivery. This study focuses on nano sizing by application of microreactor technology and nanoparticles to obtain polymeric particulate with a selection of model drugs for inhalation drug delivery routes. The development of nanoparticles of two challenging compounds in terms of solubility and permeability, namely Ibuprofen (IBU) and Salmeterol (SAL), was conducted using a continuous, controlled, and scalable system offered by microfluidic reactor with the incorporation of anti-solvent approach. The research explores the potential of this technology to enhance absorption rate and hence bioavailability of IBU via oral route, and SAL via inhalation. IBU, an anti-inflammatory drug, is classified as BCS Class II drug with low solubility and high permeability. SAL is a selective long acting β2-agonist which is co-dispensed along with a short-acting β2-agonist for quick relief of acute bronchoconstriction due to its long onset of action. This lack of the ‘kick’ effect in SAL can be attributed to its relatively higher lipophilicity which causes a delay in the diffusion to the β2 receptors on the smooth muscles. It is therefore feasible to assume that increasing the dissolution and/or diffusion rate of SAL in the interstitial fluids would reduce the delay between administration and the onset of action of this drug which would be beneficial to patients. Process and formulation parameters were investigated to optimize the production and stability of nano particles of both drugs using Y shaped microfluidic reactors. IBU results show that the smaller the angle between the two inlets were the smaller the particle size achieved. Moreover, the particle size increased with increasing the concentration of IBU solution. The effect of the polymer mixture ratio (PVP/HPMC) on the initial particle size was not clear though. The smallest particle size (113 nm) was achieved using 10° Y shaped chip with IBU concentration of 1 mg/mL and a polymer mixture of 0.3% w/v PVP and 0.5% w/v HPMC. Using a polymer mixture of 0.5% w/v of each polymer though yielded a better PDI (140nm and PDI of 0.5). Same observations were noted when the syringe pumps were replaced with a non-pulsatile pressure pump. Particle size though dropped significantly to 33nm. Stability data showed that all systems were practically stable regardless of the process or formulation parameters. In addition, a considerable 2.5 fold increase in dissolution rate was observed in the first 20 minutes when compared to the raw material. The optimized parameters were applied to SAL to produce nanocrystals with best result (59 nm) were obtained using 50µg/mL Salmeterol with microfluidics inlet angle 10° with non-pulse syringe pump. The stabilizing mixture was PVP 0.8% w/v and Tween 80 at a concentration of 0.02%. This approach offered a basis for the generation of nano sized SAL particles with higher fine particle fraction and better deposition in NGI than currently marketed formulations, thus providing a more efficient drug dose delivery and lung deposition.
-
Determinants and consequences of attribution statements on corporate financial performance outcomes in the annual report. An empirical analysis of UK listed firms.This thesis explores causal attribution statements on performance outcomes given in annual reports of UK listed rms. The objectives are three-fold. First, it analyses the nature and extent of attribution statements provided. Second, it explores corporate governance factors and rm-speci c characteristics that are related to the provision of attribution statements. Finally, it investigates the economic consequences of providing attribution statements by examining their association with the rm's cost of equity capital. Using data drawn from a sample of 142 UK rms listed on the London Stock Exchange, content analysis was used to measure the extent of attributions in the annual reports for the year 2006. The results show that the volume of attribution statement provision is generally low and variation across rms is low. Firms also show a strong tendency to explain performance with internal rather than with external reasons. The results from regression analysis show that the volume of attribution statements and the space given to internal and external attribution statements is associated with the proportion of non-executive directors, director share ownership, audit committee size, market value, gearing, pro tability and new share issues. With respect to the relationship between the attribution statements and the cost of capital, the PEG model was employed to estimate the cost of equity capital. The ndings indicate an association between attribution statement provision and the cost of equity capital, but only for rms with low analyst following. For these rms, more extensive performance explanations and more extensive internal explanations are associated with a higher cost of equity capital. However, attribution statements are unrelated to the cost of equity capital for rms with high analyst following. The thesis makes two contributions in the area of attribution determinants. First, it measures attribution provision with a measure that has not been previously applied in the literature to measure attribution statements. Second, it provides evidence on how rm-speci c characteristics and the rm's corporate governance mechanisms in uence the extent and the type of performance explanations provided by rms. The thesis makes four contributions regarding the e ect of attribution statements on the cost of capital. First, it uses a quantitative approach to directly estimate the cost of capital e ects of attribution statements. Second, it provides evidence that the association between attribution statements and the cost of equity capital is in uenced by an interaction between attribution statements and analyst following. Third, the thesis provides the rst evidence of the relationship between attribution statements and the cost of equity capital in a UK setting. Fourth, it provides evidence that the relationship between disclosure and the cost of equity capital is complex and is in uenced by interactions between disclosure and information intermediaries.
