Measuring object-oriented software maintainbility in design phase using structural complexity and aesthetic metrics

Size estimation is one of important processes related to success of software project management. This paper presents novel software size estimation model by using Multilayer Perceptron approach. Software size in terms of Lines of code is used as criterion variable. Structural complexity metrics are used as predictors. The metrics can be captured from a software design model named UML Class diagram. A high predictive ability of the model is shown with correlation coefficient measure. Moreover, four training algorithms; Levenberg-Marquardt, Scaled Conjugate Gradient, Broyden-Fletcher-Golfarb-Shanno and Bayesian Regularization, have been applied on the network for better estimation. The obtained results indicate the highest accuracy on the model with Bayesian Regularization algorithm.

Quality of data models for data warehouse has significant effect on the quality of data warehouse. Complexity metrics play significant role in predicting quality attributes of a software artifact. Few researchers have proposed structural complexity metrics for the multidimensional data models for data warehouse which may act as objective indicators of the quality of these models. However, the metrics proposed earlier have not considered the structural complexity due to relationships among various elements present in these models. This paper proposes a complexity metric which considers structural complexity due to relationships among elements present in multidimensional models for data warehouse. The metric is proposed on the basis of Goal Question Metric approach. The practical usefulness of the proposed metric is proved by validating the metric using a practical framework proposed by Kaner. This preliminary validation suggests that the metric may be linked to the quality of the multidimensional models. The advantage of the metric is that it is available during early phase of software development life cycle. The metric will also help the developers to select quality data model among various semantically equivalent models.

This study evaluates the potential of low-altitude airborne laser scanning (ALS) and terrestrial laser scanning (TLS) for characterizing structural complexity in Southern Finland. Unlike species diversity, structural complexity reflects realized niche occupancy by describing how vegetation utilizes light, water, and space, providing key insights into ecosystem functioning. We analyzed 99 circular sample plots (r = 20 m) scanned with helicopter-borne ALS at 80 m altitude and TLS data from nine scan locations per plot. Structural complexity metrics were derived at both grid level (variability in canopy height models and voxel occupancy) and object level (variability in individual tree attributes). High-density ALS effectively captured vertical and horizontal complexity through object-level analysis, showing close agreement with TLS. However, differences in measurement geometry affected volumetric complexity, with ALS and TLS characterizing tree architecture and vegetation occupancy differently. Object-level approaches captured a broader range of horizontal and vertical complexity, while grid-level approaches better captured volumetric variability, facilitating the identification of forest stand properties and biodiversity hotspots. The strongest agreement between ALS and TLS occurred for variation in tree height (R² = 0.66, Spearman = 0.80), while lowest agreement was found for fractal dimensions of tree architecture (R² = 0.04, Spearman = 0.25).

Structural complexity refers to the three-dimensional arrangement and variability of both biotic and abiotic components of an ecosystem. Metrics that characterize structural complexity are often used to manage various aspects of ecosystem function, such as light transmittance, wildlife habitat, and biological diversity. Additionally, these metrics aid in evaluating resilience to disturbance events, including hurricanes, bark-beetle outbreaks, and wildfire. Recent advances in wildland fire modelling have facilitated the integration of forest structural complexity metrics into the QUIC-Fire model, enabling real-time prediction of fire spread and behaviour by simulating interactions between fire, weather, topography, and forest structure. While QUIC-Fire is designed to be highly adaptable, model performance depends on the availability and accuracy of local data inputs. Expanding the model's usability across different regions can be facilitated by the availability of more comprehensive and high-quality data. Thus, the primary goal behind the data products we developed was to establish a basis for collaborative research across various disciplines, particularly within the focal areas of the Southern Research Station, such as forestry, wildland fire, hydrology, soil science, and cultural resources at Bent Creek, Coweeta, Escambia, and Hitchiti Experimental Forests (EFs).Airborne laser scanning (ALS) was used to collect point-cloud data for each EF during the leaf-off season to minimize interference from foliage. Subsequent processing of the raw lidar data involved outlier detection and filtering, ground and non-ground classification, and the computation of a variety of metrics representing various aspects of topography and forest structure at both the pixel-level and the tree-level. Pixel-level topographic data products include: digital elevation model (DEM), slope, aspect, topographic position index (TPI), topographic roughness index (TRI), roughness, and flow direction. Forest structural-complexity metrics include canopy height, foliar height diversity (FHD), vertical distribution ratio (VDR), canopy rugosity, crown relief ratio (CRR), understory complexity index (UCI), vertical complexity index (VCI), canopy cover, mean vegetation height, and the standard deviation of vegetation height. Tree-level data products were computed from the point cloud using multiple algorithms to perform individual tree detection (ITD) and individual tree segmentation (ITS). The datasets have been harmonized and are openly accessible through the USDA Forest Service Research Data Archive.

BackgroundModern theories define chronic pain as a multidimensional experience – the result of complex interplay between physiological and psychological factors with significant impact on patients' physical, emotional and social functioning. The development of reliable assessment tools capable of capturing the multidimensional impact of chronic pain has challenged the medical community for decades. A number of validated tools are currently used in clinical practice however they all rely on self-reporting and are therefore inherently subjective. In this study we show that a comprehensive analysis of physical activity (PA) under real life conditions may capture behavioral aspects that may reflect physical and emotional functioning.MethodologyPA was monitored during five consecutive days in 60 chronic pain patients and 15 pain-free healthy subjects. To analyze the various aspects of pain-related activity behaviors we defined the concept of PA ‘barcoding’. The main idea was to combine different features of PA (type, intensity, duration) to define various PA states. The temporal sequence of different states was visualized as a ‘barcode’ which indicated that significant information about daily activity can be contained in the amount and variety of PA states, and in the temporal structure of sequence. This information was quantified using complementary measures such as structural complexity metrics (information and sample entropy, Lempel-Ziv complexity), time spent in PA states, and two composite scores, which integrate all measures. The reliability of these measures to characterize chronic pain conditions was assessed by comparing groups of subjects with clinically different pain intensity.ConclusionThe defined measures of PA showed good discriminative features. The results suggest that significant information about pain-related functional limitations is captured by the structural complexity of PA barcodes, which decreases when the intensity of pain increases. We conclude that a comprehensive analysis of daily-life PA can provide an objective appraisal of the intensity of pain.

Complexity, interpretability and robustness of GP-based feature engineering in remote sensing

Provides a toolkit for calculating forest and canopy structural complexity metrics from terrestrial LiDAR (light detection and ranging).

This chapter handles assessment of the innermost layer of the Semantic Web cake, i.e., ontology, through a metrics suite called OntoMod. We assess the syntactical composition of ontology, in addition to its behavioral aspect in a Semantic Web application. Firstly, novel cohesion and coupling metrics have been discussed, which are found to be superior in terms of accuracy and expressive power than existing cohesion and coupling metrics, on sample ontologies. Furthermore, by using continuous scale of measurement, and tackling indirect relationships in ontology, our cohesion and coupling metrics can account for subtle dissimilarities among different modularizations of the same ontology. Hence, a key role of OntoMod is to help the ontology engineer in choosing the most suitable modularization of ontology according to fitness for (re)use, regardless of its type and size. Secondly, improved metrics for structural complexity of an ontology have been formulated. Finally, a few crucial metrics for behavioral assessment of ontology have been adapted from the literature and incorporated in the OntoMod suite. The only precondition for assessing structural and behavioral quality of modular ontology through the OntoMod suite is the availability of partial or complete ontology visualization. This chapter validates each of the constituent OntoMod metrics, either practically on real-world ontologies or theoretically through benchmark validation frameworks.

객체지향 시스템의 복잡성 척도에 대하여 많은 연구와 검증이 이루어져 왔다. 대부분의 척도들은 시스템의 부분적 측면 예를 들어, 객체 간 결합도, 상속 구조의 복잡도, 메소드의 응집도 등에 대한 측정을 목표로 하고 있다. 그런데 소프트웨어 실무자들은 부분적이 아닌 시스템의 전반적인 복잡도를 측정할 수 있기를 바라고 있다. 본 논문은 UML의 클래스 다이어그램을 분석함으로써 객체지향 시스템의 전체적 구조에 대한 복잡도를 연구한 것이다. 클래스 다이어그램은 클래스와 클래스 간 관계로 구성되어 있다. 관계에는 연관 관계, 일반화 관계, 집합 관계 둥 세 가지가 있는데, 이 관계들이 객체지향 시스템의 구조를 이해하기 어렵게 하고 있다. 본 연구에서는 이 세 가지 관계를 통합하여 객체지향 시스템의 복잡도를 측정하는 경험적 척도를 제안하고 있다. 이 척도는 소프트웨어 개발자가 코딩하기 전에 객체지향 시스템의 복잡도를 평가해 보고 필요시 설계를 수정할 수 있도록 하게 함으로써 설계 업무에 많은 도움을 줄 것이다. Many researches and validations for the complexity metrics of the object-oriented systems have been studied. Most of them are aimed for the measurement of the partial aspects of the systems, for example, the coupling between objects, the complexity of inheritance structures, the cohesion of methods, and so on. But the software practitioners want to measure the complexity of overall system, not partial. We studied the complexity of the overall structures of object-oriented systems by analyzing the class diagram of UML. The class diagram is composed of classes and their relations. There are three kinds of relations, association, generalization, and aggregation, which are making the structure of object-oriented systems to be difficult to understand. We proposed a heuristic metric to measure the complexity of object-oriented systems by putting together the three kinds of the relations. Tn analyze the complexity of the structure of a object-oriented system for the maintainability of the system, we measured the degree of understandability of it, the reverse engineering time to draw a class diagram from the source codes, and the number of errors in the diagram. The results of this experiment shows that our proposed metric has a considerable relationship with the complexity of object-oriented systems. The metric will be helpful to the software developers for their designing tasks by evaluating the complexity of the structures of object-oriented systems and redesigning tasks , of them for the future maintainability.

A novel metric for assessing structural complexity of data warehouse requirements models

A methodology for generating an optimum software quality classification tree using software complexity metrics to discriminate between high-quality modules and low-quality modules is proposed. The process of tree generation is an application of the AIC (Akaike Information Criterion) procedures to the binomial distribution. AIC procedures are based on maximum likelihood estimation and the least number of complexity metrics. It is an improvement of the software quality classification tree generation method proposed by Porter and Selby (1990) from the viewpoint that the complexity metrics are minimized. The problems of their method are that the software quality prediction model is unstable because it reflects observational errors in real data too much and there is no objective criterion for determining whether the discrimination is appropriate or not at a deep nesting level of the classification tree when the number of sample modules gets smaller. To solve these problems a new metric is introduced and its validity is theoretically and experimentally verified. In our examples, complexity metrics written in C language, such as lines of source code, Halstead's (1977) software science, McCabe's (976) cyclomatic number, Henry and Kafura's (1981) fan-in/out and Howatt and Baker's (1989) scope number, are investigated. Our experiments with a medium-sized piece of software (85 thousand lines of source code; 562 samples) show that the software quality classification tree generated by our new metric identifies the target class of the observed modules more efficiently using the minimum number of complexity metrics without any significant decrease of the correct classification ratio (76%->72%) than the conventional classification tree.

This paper experimentally evaluates the usefulness of a business application framework from a viewpoint of saving cost and quality of the software in a company. Here, we conducted two case studies. In the case studies, four kinds of applications are developed. Each of them is developed in two ways: using framework-based reuse and conventional module-based reuse. Then, we evaluate the difference among them using the several functionality and complexity metrics. As the results, the framework-based reuse would be more efficient than the module-based reuse in the company.

Structurally complex forests optimize resources to assimilate carbon more effectively, leading to higher productivity. Information obtained from Light Detection and Ranging (LiDAR)‐derived canopy structural complexity (CSC) metrics across spatial scales serves as a powerful indicator of ecosystem‐scale functions such as gross primary productivity (GPP). However, our understanding of mechanistic links between forest structure and function, and the impact of disturbance on the relationship, is limited. Here, we paired eddy covariance measurements of carbon and water fluxes from nine forested sites within the 10 × 10 km CHEESEHEAD19 study domain in Northern Wisconsin, USA with drone LiDAR measurements of CSC to establish which CSC metrics were strong drivers of GPP, and tested potential mediators of the relationship. Mechanistic relationships were inspected at five resolutions (0.25, 2, 10, 25, and 50 m) to determine whether relationships persisted with scale. Vertical heterogeneity metrics were the most influential in predicting productivity for forests with a significant degree of heterogeneity in management, forest type, and species composition. CSC metrics included in the structure‐function relationship as well as driver strength was dependent on metric calculation resolution. The relationship was mediated by light use efficiency (LUE) and water use efficiency (WUE), with WUE being a stronger mediator and driver of GPP. These findings allow us to improve representation in ecosystem models of how CSC impacts light and water‐sensitive processes, and ultimately GPP. Improved models enhance our capacity to accurately simulate forest responses to management, furthering our ability to assess climate mitigation strategies.

In the early stage of software development, design complexity metrics are considered as useful indicators of a software testing effort and quality attributes. However, existing works made great efforts in establishing standardized metrics to evaluate the complexity of software, but there have not been significant efforts in finding the correlations among the cognitive complexitymetrics. To address this challenge, this paper reviewed cognitive complexity metrics which includes: Improved Cognitive Complexity Measure (ICCM), New Cognitive Complexity of Program (NCCoP) and Modified Cognitive Complexity Measure (MCCM). The metrics were employed to analyse some selected sorting algorithms implemented in a procedural C programming language. The relationships among the aforementioned metrics were calculated using the Pearson Correlation Coefficient Method. The results of the comparative examination of ICCM, NCCoP and MCCM revealed that ICCM had more responsive measurements and that there exists a strong relationship among the specified metrics. ICCM had the strongest significance among the considered metrics based on the efforts in comprehending the information contained in the sorting algorithm codes. The study contributed significantly to understanding and addressing the complexity emanating from software development.

The research topic focuses on the characterization of software quality considering the main software elements such as people, process and product. Many attributes (size, language, testing techniques etc.) probably could have an effect on the quality of software. In this thesis we aim to understand the impact of attributes of three P's (people, product, process) on the quality of software by empirical means. Software quality can be interpreted in many ways, such as customer satisfaction, stability and defects etc. In this thesis we adopt ‘defect density' as a quality measure. Therefore the research focus on the empirical evidences of the impact of attributes of the three P's on the software defect density. For this reason empirical research methods (systematic literature reviews, case studies, and interviews) are utilized to collect empirical evidence. Each of this research method helps to extract the empirical evidences of the object under study and for data analysis statistical methods are used. Considering the product attributes, we have studied the size, language, development mode, age, complexity, module structure, module dependency, and module quality and their impact on project quality. Considering the process attributes, we have studied the process maturity and structure, and their impact on the project quality. Considering the people attributes, we have studied the experience and capability, and their impact on the project quality. Moreover, in the process category, we have studied the impact of one testing approach called ‘exploratory testing' and its impact on the quality of software. Exploratory testing is a widely used software-testing practice and means simultaneous learning, test design, and test execution. We have analyzed the exploratory testing weaknesses, and proposed a hybrid testing approach in an attempt to improve the quality. Concerning the product attributes, we found that there exist a significant difference of quality between open and close source projects, java and C projects, and large and small projects. Very small and defect free modules have impact on the software quality. Different complexity metrics have different impact on the software quality considering the size. Product complexity as defined in Table 53 has partial impact on the software quality. However software age and module dependencies are not factor to characterize the software quality. Concerning the people attributes, we found that platform experience, application experience and language and tool experience have significant impact on the software quality. Regarding the capability we found that programmer capability has partial impact on the software quality where as analyst capability has no impact on the software quality. Concerning process attributes we found that there is no difference of quality between the project developed under CMMI and those that are not developed under CMMI. Regarding the CMMI levels there is difference of software quality particularly between CMMI level 1 and CMMI level 3. Comparing different process types we found that hybrid projects are of better quality than waterfall projects. Process maturity defined by (SEI-CMM) has partial impact on the software quality. Concerning exploratory testing, we found that exploratory testing weaknesses induce the testing technical debt therefore a process is defined in conjunction with the scripted testing in an attempt to reduce the associated technical debt of exploratory testing. The findings are useful for both researchers and practitioners to evaluate their projects