In agile software development, product backlog items (PBI) are used to capture the user requirements prior to the product implementation. Many types of requirements can be observed within a software project. Proper classification of PBI can positively impact the software development process. PBI can be classified into three categories: user stories, foundational stories, and spikes. After the extreme literature survey, no research was held on classifying the PBI into the categories mentioned above. This paper proposed a machine learning (ML) based approach to classify the PBI into three categories. 4,721 PBI were collected from different software projects and manually labelled into the three classes mentioned above. Then the PBI were cleaned using different pre-processing techniques. Classification models were constructed using ML techniques. The performance of each ML model was evaluated using accuracy, precision, recall, and F1 score. Support vector machine (SVM) outperformed other ML models by providing 88% accuracy.

Crime is a social and economic problem that affects a country's quality of day-to-day life and economic growth. However, analyzing and forecasting crime is not a straightforward job for a law enforcement investigator to manually unravel the underlying nuances of crime data. To make this process easier and more automated, the authors present a machine-learning model for crime analysis and predictions. The authors used a London crime dataset and enhanced the data set by incorporating population density, percentage of economically inactive working age, and average monthly temperature. The pre-process step prepares the raw data and makes it suitable for the machine-learning model. Bagging and boosting ensemble techniques were used to find a better- machine-learning model. GridSearchCV was used to tune hyperparameters to find the best-performed model. Parameters were tuned as an iterative processes. Eventually, the researchers compared all the algorithms and selected the Random Forest bagging regression model as the best-performed algorithm.

Road accidents, causing 1.35 billion deaths and 50 million injuries annually, are a significant global issue that demands timely detection and prevention. This study reviews existing research on road accident detection using data mining techniques. In this research, the authors developed a method for classifying road accident-related tweets using Twitter mining. They collected a dataset of road accident-related tweets, pre-processed them, and cleaned the data using natural language processing. Various machine learning models were applied to classify tweets into real-time, traffic, and informative categories, including SVM, logistic regression, ANN, LSTM with TF-IDF, and LSTM with BERT. The LSTM model with BERT exhibited the highest precision and recall scores of 0.88 and 0.87, respectively. The findings highlight the potential of Twitter mining for real-time road accident detection. Despite model accuracy and robustness limitations, this research is a promising starting point for leveraging social media data to enhance road safety.