Technology mancation Towards an Automatic Linguistic Quality Assurance of Localized Software and Website based on Multidimensional Quality Metrics

Context: Nowadays, the main attempt of engineers and scholars is to automate the whole or some phases of a process, particularly in the field of software design and development. Testing localized software is also a good area for automation. While there are studies on the topic, however, the available literature does not propose an automation tool on the basis of the well-known error topologies of available quality control models, namely, LISA QA Mode, SAE J۲۴۵۰ and Multidimensional Quality Metrics (MQM). Objectives: In this research, the objective is to propose an approach or tool which can be used during software testing, particularly after localization process to get insight into the linguistic quality state of a localized software or a website. Methods: To this end, seemingly, machine learning algorithms, such as Decision Trees Random Forests, Support Vector Machines (SVM), Neural Networks (e.g., Convolutional Neural Networks, Recurrent Neural Networks) and K-Nearest Neighbors (KNN) can be employed. Results: After careful consideration of the available literature, and research articles indexed in WoS, Scopus, Science Direct, IEEE Xplore and ACM Digital Library, it was found that there is no systematic approach or a tool for automatic assessment of localized software based on MQM metrics. Conclusion: In conclusion, we need a tool which is capable of automatic quality assurance of localized software, but it is highly critical to develop this tool based on de facto quality standards of software localization industry.

Keywords:

Linguistic Quality Assurance , MQM Metrics , Software Localization , Automatic Testing