Negative results for software effort estimation

More than half the literature on software effort estimation (SEE) focuses on comparisons of new estimation methods. Surprisingly, there are no studies comparing state of the art latest methods with decades-old approaches. Accordingly, this paper takes five steps to check if new SEE methods generated better estimates than older methods. Firstly, collect effort estimation methods ranging from “classical” COCOMO (parametric estimation over a pre-determined set of attributes) to “modern” (reasoning via analogy using spectral-based clustering plus instance and feature selection, and a recent “baseline method” proposed in ACM Transactions on Software Engineering). Secondly, catalog the list of objections that lead to the development of post-COCOMO estimation methods. Thirdly, characterize each of those objections as a comparison between newer and older estimation methods. Fourthly, using four COCOMO-style data sets (from 1991, 2000, 2005, 2010) and run those comparisons experiments. Fifthly, compare the performance of the different estimators using a Scott-Knott procedure using (i) the A12 effect size to rule out “small” differences and (ii) a 99 % confident bootstrap procedure to check for statistically different groupings of treatments. The major negative result of this paper is that for the COCOMO data sets, nothing we studied did any better than Boehms original procedure. Hence, we conclude that when COCOMO-style attributes are available, we strongly recommend (i) using that data and (ii) use COCOMO to generate predictions. We say this since the experiments of this paper show that, at least for effort estimation, how data is collected is more important than what learner is applied to that data.     

A parametric bootstrap approach for ANOVA with unequal variances: Fixed and random models

This article is about testing the equality of several normal means when the variances are unknown and arbitrary, i.e., the set up of the one-way ANOVA. Even though several tests are available in the literature, none of them perform well in terms of Type I error probability under various sample size and parameter combinations. In fact, Type I errors can be highly inflated for some of the commonly used tests; a serious issue that appears to have been overlooked. We propose a parametric bootstrap (PB) approach and compare it with three existing location-scale invariant tests—the Welch test, the James test and the generalized F (GF) test. The Type I error rates and powers of the tests are evaluated using Monte Carlo simulation. Our studies show that the PB test is the best among the four tests with respect to Type I error rates. The PB test performs very satisfactorily even for small samples while the Welch test and the GF test exhibit poor Type I error properties when the sample sizes are small and/or the number of means to be compared is moderate to large. The James test performs better than the Welch test and the GF test. It is also noted that the same tests can be used to test the significance of the random effect variance component in a one-way random model under unequal error variances. Such models are widely used to analyze data from inter-laboratory studies. The methods are illustrated using some examples.     

Differences in prepulse inhibition (PPI) between Wistar and Sprague-Dawley rats clarified by a new method of PPI standardization.

Rat strain differences in the acoustic startle response (ASR) and prepulse inhibition (PPI) of that response are of increasing interest, especially as the genetics of PPI may provide an approach to studying the genetics of certain mental illnesses. However, strain differences in PPI are confounded by differences in ASR. To clarify this issue, the authors investigated the ASR and PPI across a range of startling stimulus intensities (70 dB-120 dB) in Wistar and Sprague-Dawley rats (N=96). Sprague-Dawleys showed more PPI of ASR capacity (response limit) than Wistars. In contrast, Wistars exhibited greater PPI than Sprague-Dawleys, as measured by an increase in response threshold. This dissociation suggests that PPI is more complex than that assessed by single startling stimulus intensity. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

High‐performance electromagnetic interference shielding material based on an effective mixing protocol

A facile and economic method is developed for the fabrication of new lightweight materials with high electromagnetic interference (EMI) shielding performance, good mechanical properties and low electrical percolation threshold through melt mixing. Electrical properties, DC conductivity, EMI shielding performance and mechanical properties of poly(trimethylene terephthalate) (PTT)/multiwalled carbon nanotube (MWCNT) nanocomposites with varying filler loading of MWCNTs were investigated. High‐resolution transmission electron microscopy was used to determine the distribution of MWCNTs in the PTT matrix. The newly developed nanocomposites show excellent dielectric and EMI shielding properties. Theoretical electrical percolation threshold was achieved at 0.21 wt% loading of MWCNTs, due to the high aspect ratio and the three‐dimensional network formation of MWCNTs. Experimental DC conductivity values were compared with those of theoretical models such as the Voet, Bueche and Scarisbrick models, which showed good agreement. The PTT/3% MWCNT composite showed an EMI shielding value of ～38 dB (99.99% attenuation) with a sample thickness of 2 mm. Power balance was used to determine the actual contribution of reflection, absorption and transmission loss to the total EMI shielding value. The nanocomposites showed good tensile and impact properties and the composite with 2% MWCNTs exhibited an improvement in tensile strength of as much as 96%. © 2018 Society of Chemical Industry     

Optimisation of maintenance strategy for rural road network using genetic algorithm

The objective of this paper is to develop an optimised maintenance strategy for the rural road network of Kerala state. This is accomplished with the development of a bi-objective deterministic optimisation model which simultaneously satisfies the objectives of both minimisation of total maintenance cost and maximisation of performance of the road network. The model is capable of planning the maintenance activities over a multi-year planning period. The performance of the road network is accounted using the composite index, namely Pavement Condition Index. The constraint-based genetic algorithm was used as the optimisation tool since it very well takes care of the combinatorial nature of the network-level pavement maintenance programming. The applicability of the model is illustrated using a case study for the rural road network of Kerala state in India. The effort made through this research work to develop a suitable Pavement Maintenance and Management System for rural road network can lead to the proper maintenance and upkeep of the rural roads, thereby triggering a positive impact on the Indian economy.     

Coercivity Control of Variable-Length Iron Chains in Phthalocyanine Thin Films

The magnetic characteristics of iron phthalocyanine thin films are studied with a vibrating sample magnetometer, identifying a ferromagnetic transition temperature at 4.5?K. The metal ions at the center of the molecule are self-assembled along chains producing quasi one-dimensional magnetic chains of variable length in the thin films. The average chain length is varied from 20 to 300?nm via substrate temperature during deposition. Below the critical transition temperature, the magnetization curves have the shape of wasp-waisted or constricted loops. The in-plane chain length modulates the coercivity and saturation field and larger grains increase the coercivity significantly. First-order reversal curves of the wasp-waisted hysteresis loops reveal a long narrow strip that suggests a broad distribution of coercive fields and weak intergrain magnetic interactions. These findings are also supported through simulations based on the Preisach model.     

Effect of diethyl ether blended with neem oil methyl esters in CI engine

Vegetable oils are now considered to be a potential alternative that can be used in place of partial or total substitution of diesel fuels. In this study, we used diethyl ether as an oxygenated additive to investigate the possible use of higher percentages of biodiesel in an unmodified diesel engine. Neem oil was selected for biodiesel production. The tests were performed at a steady-state condition in a single-cylinder constant speed DI diesel engine. The combustion process involved in diesel engine would be improved and the particulate matter would be reduced if these biodiesels are blended. An experimental investigation is carried out to establish the emission characteristics of a direct injection diesel engine using diethyl ether as additive in Neat Neem oil Biodiesel. Emissions of hydro carbon oxides of nitrogen and carbon monoxide significantly reduced by adding diethyl ether into neem oil biodiesel at 10% and 20% on volume basis.     

A characterization of partial blocks in weighted graphs

Some of the classical connectivity concepts in Graph theory are generalized in this article. Strong and strongest strong cycles are introduced. Partial blocks are characterized using strongest paths. Some necessary and sufficient conditions for a weighted graph to be a partial block are also presented.     

Azaaurones as Potent Antimycobacterial Agents Active against MDR- and XDR-TB

Herein we report the screening of a small library of aurones and their isosteric counterparts, azaaurones and N-acetylazaaurones, against Mycobacterium tuberculosis. Aurones were found to be inactive at 20 μm , whereas azaaurones and N-acetylazaaurones emerged as the most potent compounds, with nine derivatives displaying MIC₉₉ values ranging from 0.4 to 2.0 μm . In addition, several N-acetylazaaurones were found to be active against multidrug-resistant (MDR) and extensively drug-resistant (XDR) clinical M. tuberculosis isolates. The antimycobacterial mechanism of action of these compounds remains to be determined; however, a preliminary mechanistic study confirmed that they do not inhibit the mycobacterial cytochrome bc1 complex. Additionally, microsomal metabolic stability and metabolite identification studies revealed that N-acetylazaaurones are deacetylated to their azaaurone counterparts. Overall, these results demonstrate that azaaurones and their N-acetyl counterparts represent a new entry in the toolbox of chemotypes capable of inhibiting M. tuberculosis growth.