Improvement and evaluation of intellectual productivity model based on work state transition

Author(s):  
Kazune Miyagi ◽  
Shou Kawano ◽  
Hirotake Ishii ◽  
Hiroshi Shimoda
2018 ◽  
Vol 21 (1) ◽  
Author(s):  
Constanza Pérez ◽  
Beatriz Marín

[Context] The growing demand for high-quality software has caused the industry to incorporate processes to enable them to comply with these standards, but increasing the cost of development. A strategy to reduce this cost is to incorporate quality evaluations from early stages of software development. A technique that facilitates this evaluation is the model-based testing, which allows to generate test cases at early phases using as input the conceptual models of the system. [Objective] In this paper, we introduce TCGen, a tool that enables the automatic generation of abstract test cases starting from UML conceptual models. [Method] The design and implementation of TCGen, a technique that applies different testing criteria to class diagrams and state transition diagrams to generates test cases, is presented as a model-based testing approach. To do that, TCGen uses UML models, which are widely used at industry and a set of algorithms that recognize the concepts in the models in order to generate abstract test cases. [Results] An exploratory experimental evaluation has been performed to compare the TCGen tool with traditional testing. [Conclusions] Even though the exploratory evaluation shows promising results, it is necessary to perform more empirical evaluations in order to generalize the results. Abstract (in Spanish): [Contexto] La creciente demanda de software de alta calidad ha provocado que la industria incorpore procesos para permitirles cumplir con estos estándares, pero aumentando el costo del desarrollo. Una estrategia para reducir este costo es incorporar evaluaciones de calidad desde las primeras etapas del desarrollo del software. Una técnica que facilita esta evaluación es la prueba basada en modelos, que permite generar casos de prueba en fases tempranas utilizando como entrada los modelos conceptuales del sistema. [Objetivo] En este artículo, presentamos TCGen, una herramienta que permite la generación automática de casos de pruebas abstractas a partir de modelos conceptuales UML. [Método] El diseño e implementación de TCGen, una técnica que aplica diferentes criterios de prueba a los diagramas de clases y diagramas de transición de estados para generar casos de prueba, se presenta como un enfoque de prueba basado en modelos. Para hacer eso, TCGen utiliza modelos UML, que son ampliamente utilizados en la industria y un conjunto de algoritmos que reconocen los conceptos en los modelos para generar casos de prueba abstractos. [Resultados] Se realizó una evaluación experimental exploratoria para comparar la herramienta TCGen con las pruebas tradicionales. [Conclusiones] Aunque la evaluación exploratoria muestra resultados prometedores, es necesario realizar más evaluaciones empíricas para generalizar los resultados.  


2020 ◽  
Author(s):  
Haiming Wu ◽  
Ruigang Wang ◽  
Lixia Jia ◽  
Likui Feng ◽  
Xu Zhou

Abstract Social network has gradually become the mainstream way for people to obtain and interact with information. The study on the law of information dissemination in social networks is of great significance to enterprise marketing, public opinion control and social recommendation. This paper puts forward a method that use multi-dimensional node influence and epidemic model to illustrate the causes and rules of information dissemination in social networks. Firstly, based on the multiple linear regression model, a measurement method of node influence is proposed from three dimensions: topology, user interaction behavior and information content. Then, taking the node influence as the cause of state transition, the information dissemination model based on the epidemic model is constructed, and the multidimensional factors affecting the information dissemination are analyzed. Meanwhile, the information dissemination trend in social networks is described.


2020 ◽  
Vol 109 (5) ◽  
pp. 939-972
Author(s):  
Yu Nishiyama ◽  
Motonobu Kanagawa ◽  
Arthur Gretton ◽  
Kenji Fukumizu

AbstractKernel Bayesian inference is a principled approach to nonparametric inference in probabilistic graphical models, where probabilistic relationships between variables are learned from data in a nonparametric manner. Various algorithms of kernel Bayesian inference have been developed by combining kernelized basic probabilistic operations such as the kernel sum rule and kernel Bayes’ rule. However, the current framework is fully nonparametric, and it does not allow a user to flexibly combine nonparametric and model-based inferences. This is inefficient when there are good probabilistic models (or simulation models) available for some parts of a graphical model; this is in particular true in scientific fields where “models” are the central topic of study. Our contribution in this paper is to introduce a novel approach, termed the model-based kernel sum rule (Mb-KSR), to combine a probabilistic model and kernel Bayesian inference. By combining the Mb-KSR with the existing kernelized probabilistic rules, one can develop various algorithms for hybrid (i.e., nonparametric and model-based) inferences. As an illustrative example, we consider Bayesian filtering in a state space model, where typically there exists an accurate probabilistic model for the state transition process. We propose a novel filtering method that combines model-based inference for the state transition process and data-driven, nonparametric inference for the observation generating process. We empirically validate our approach with synthetic and real-data experiments, the latter being the problem of vision-based mobile robot localization in robotics, which illustrates the effectiveness of the proposed hybrid approach.


Sign in / Sign up

Export Citation Format

Share Document