homefy — The intelligent way of choosing the right location to study
This project was carried out as part of the TechLabs “Digital Shaper Program” in Münster (summer term 2021).
The homefy program offers its users a solution to find out which city best suits their personality and desired interests. It solves the widespread problem that people, who want to move to a new city usually don’t know whether the desired city corresponds their imagination, as well as whether there is a match between the user’s individual lifestyle and the new place of residence. We have focused on the target group of students who, after graduating from high school, vocational training or their first years at work, would like to take the step towards studying, which often requires the students to move within Germany. The program scrapes all relevant information from the internet, makes it comparable and defines scores, which when multiplied with the users input determines the favorite city. It can be accessed through an user-friendly interface.
Deciding on the right place to study coupled with personal requirements and desires can be a tricky business. When looking for a place to live, everyone has probably sat in front of a computer and sourced apartment-hunting websites and the usual shared apartment portals without being able to make an informed decision about which city is actually best suited to their personal interests and needs. The difficulty lies in grasping and comparing the density of the housing market jungle coupled with the large selection of study programs, as well as one’s financial possibility. Therefore, we developed the homefy program to help students make a more informed decision when choosing a new city and hopefully loved future home. The app allows students to transparently enter their interests, wishes and their importance for their new place of residence into the mask and thus access a decision-making aid for the start of their studies. The goal of our platform is to find a beloved future home for everyone.
After the brainstorming phase, as a group we came up with criteria for factors that might play a decisive role for individuals when choosing a city to study in. We consider important the amount of rent, housing availability, choice of study programs, sports clubs, pubs, clubs & bars, gyms & swimming pools, as well as the percentage of students and the size of the city. In the next step, we searched out the appropriate sources for the criteria and scraped this information with the library “beautiful soup”. The important thing here was that the scraping had to be done for each source page again. Afterwards, a list of the five largest student cities in North Rhine-Westphalia was created using Excel. These cities are Münster, Düsseldorf, Cologne, Bochum and Aachen. After that, a data frame was created with the previously defined criteria. The number of hits of criteria on locations were cut in the next step and resulted in a series, which was added to the data frame. It is worth mentioning that we divided our criteria into soft and hard criteria. The hard criteria (city size/proportion of students in the total population) do not change over a certain period of time, they are static.
All these functions are stored in a CSV-file and can be accessed by the user through a corresponding front-end, which was created in the form of a graphical user interface (GUI). For this we used the library “tkinter”. With the help of this program, it is possible to generate labels, i.e. text modules, as well as so-called radiobuttons, which are displayed to the user and can be clicked on by the user. Thus, the labels were used to ask the user the questions concerning the relevance and preference of the previously defined criteria. To answer the questions, the user is provided with the buttons just mentioned, which are arranged on a scale from “irrelevant” to “very relevant”. Based on the selected relevance of the individual criteria, these are included to a lesser or greater extent in the overall evaluation of the respective study cities. The arrangement of the labels and buttons must be in the form of a table. Figure 1 shows a section of the user interface. Finally, the back-end was connected to the front-end.
Results of the project
First of all, the collected data provides an overview of the five largest study locations in NRW and thus also represents a well-structured information function. Above all, however, the collected information, individually adapted to one’s own preferences, can be incorporated into the decision and thus lead to a more informed decision regarding the place of study. The program gives students the city that best suits them as well as the corresponding score. The score can then also be used to evaluate how well the city ultimately fits the person and his or her preferences, or whether it is merely the best fit among the five cities. In addition, the key facts of the city are displayed and the next most suitable cities are listed in a ranking.
The application can help students to feel more comfortable with their chosen study location based on an objective decision up front and make it a better fit for them, which could ultimately reduce dropouts.
On the basis of this project, the written code and the developed system could now be extended to other cities in NRW or to other federal states, to enhance the overall fit between the place of study and personal preferences and their importance.
Kai Klöcker: Data Science: Python
Daniel Barkowski Data Science: Python
Fabienne Pollinger Data Science: Python
Charlotte Fränkel Data Science: Python