Context. Reports suggest that defects in code cost the US in excess of $50billion per year to put right. Defect Prediction is an important part of Software Engineering. It allows developers to prioritise the code that needs to be inspected when trying to reduce the number of defects in code. A small change in the number of defects found will have a significant impact on the cost of producing software. Aims. The aim of this dissertation is to investigate the factors which a ect the performance of defect prediction models. Identifying the causes of variation in the way that variables are computed should help to improve the precision of defect prediction models and hence improve the cost e ectiveness of defect prediction. Methods. This dissertation is by published work. The first three papers examine variation in the independent variables (code metrics) and the dependent variable (number/location of defects). The fourth and fifth papers investigate the e ect that di erent learners and datasets have on the predictive performance of defect prediction models. The final paper investigates the reported use of di erent machine learning approaches in studies published between 2000 and 2010. Results. The first and second papers show that independent variables are sensitive to the measurement protocol used, this suggests that the way data is collected a ects the performance of defect prediction. The third paper shows that dependent variable data may be untrustworthy as there is no reliable method for labelling a unit of code as defective or not. The fourth and fifth papers show that the dataset and learner used when producing defect prediction models have an e ect on the performance of the models. The final paper shows that the approaches used by researchers to build defect prediction models is variable, with good practices being ignored in many papers. Conclusions. The measurement protocols for independent and dependent variables used for defect prediction need to be clearly described so that results can be compared like with like. It is possible that the predictive results of one research group have a higher performance value than another research group because of the way that they calculated the metrics rather than the method of building the model used to predict the defect prone modules. The machine learning approaches used by researchers need to be clearly reported in order to be able to improve the quality of defect prediction studies and allow a larger corpus of reliable results to be gathered.

Цель работы – разработка модели машинного обучения для автоматического сопоставления описаний продуктов, представленных в текстовом виде с внутренними кодами компании. Объект исследования – бизнес-процесс сопоставления описаний продуктов с внутренними кодами компании. Методы исследования: предварительная обработка данных, анализ данных, выбор и обучение модели машинного обучения, оценка производительности модели. Результаты работы: разработана и обучена модель машинного обучения на основе алгоритма CatBoost для автоматического сопоставления описаний продуктов с внутренними кодами компании. Модель показала высокую точность и полноту при тестировании. Созданная модель машинного обучения внедрена в продуктивное использование компании АО «Сони Электроникс» и позволяет сокращать ресурсы аналитиков в существенном объеме. Выпускная квалификационная работа выполнена в текстовом редакторе Microsoft Word и представлена в электронном и печатном виде. / The goal of the work is to develop a machine learning model for automatically comparing product descriptions presented in text form with the company’s internal codes. The object of study is the business process of comparing product descriptions with internal company codes. Research methods: data preprocessing, data analysis, selection and training of a machine learning model, evaluation of model performance. Results of the work: a machine learning model based on the CatBoost algorithm was developed and trained to automatically compare product descriptions with internal company codes. The model showed high accuracy and completeness during testing. The created machine learning model has been put into productive use by Sony Electronics JSC and makes it possible to reduce analyst resources to a significant extent. The final qualifying work was completed in the text editor Microsoft Word and presented in electronic and printed form.