This is an inter-track project which involves the UI/UX team, Data Science team and Sodtware Engineering team. As part of the data science team, our task is to scrape the necessary data and then develop a machine learning model to help house seekers in Lagos have an idea of how much it would cost them to rent an apartment in locations around Lagos.
In this project, automatic text summarisation was explored BART and T5 models. Both models perform well on a variety of tasks out-of-the-box such as translation, questions answering and summarisation. The aim here, however, was to test their effectiveness on text summarisation. The models gave very good outputs which can be primed further by changing the pre-train files and hyperparameter tuning.
The purpose of this project is to develop a clustering algorithm to help a supermarket owner create target groups for his customers. This would help him know what age to target which product. As well as what kind of offers to use on which customers during his marketing runs. This is an unsupervised learning problem hence we used Kmeans clustering to segment the customers and provide Mr Ken with the Target Customers.
A classification project to reduce the tedious process of ascertaining if a person will be offered loan or not. Models used are, LogisticRegression, DecisionTreeClassifier, RandomForestClassifier and AdaBoostClassifier. While AdaBoostClassifier outshined the others further efforts were made to improve on the accuracy of the models.
Rent data was scraped from nigeriapropertycentre using Python and BeautifulSoup and were preprocessed in Excel to inspect the data for irregularities. EDA was then carried out on the data using python in jupyter-lab environment. A LinearRegression mdoel was then built to predict the rental price of houses within Lagos.
Exploratory Data Analysis is an important step required to complete every project regardless of type of data one is working with. It gives us a sense of what additional work should be performed to quantify and extract insights from our data. In this analysis, I will be looking at the different factors that could have affected the survival rate of the passengers of the Titanic.