Smart cities approach does not only emphasize the implementations of new technologies in a city but also highlights the importance of using new technologies for enabling citizens’ engagement in urban planning processes. In that regard, ICTs play a vital role in (i) supporting citizens to report their complaints related to the public spaces (i.e. via smartphone apps and social media) and (ii) providing information to citizens when and where problems occur in the city (i.e. open data platforms and dashboards). Public space complaints indicate problems related to the condition of streets and facilities in public spaces.
The collection of data on complaints results in large amounts of data which can be used for understanding the current status and making predictions for future complaints. However, in the literature and in the practice, this topic has not been investigated in-depth yet since such data on civic complaints are new. Nevertheless, Customer Journey Optimization solutions have emerged in other fields like Process and Data Mining that they can offer proper solutions for predicting the next interactions of customers (here citizens) based on the previous context of their journeys (here the sequences of their public space complaints).
In this master project, the aim is three-fold:
- to discover the process model of citizens’ behaviors using the timestamped touchpoints between Eindhoven city citizens and the public space;
- to explore the Eindhoven city data on citizens’ public space complaints, via unsupervised methods (e.g. spatial clustering techniques, spatial outlier detection methods …etc.), for understanding the inter-dependency between complaints in each cluster and finding clusters with similar complaint behaviors or the correlation between outliers on specific attributes. This validity of the findings in this part will be examined through a close support from the domain knowledge (semi-supervised learning);
- to develop a model to predict the occurrence of citizens’ public space complaints in different clusters and in an online matter. To improve the prediction accuracy, this should be performed using the minimal subset of correlated attributes.
In this master project, the data of more than 350,000 public citizen complaints from Eindhoven will be used (collected between 2013 and 2020). An example visualization of the data can be seen in Figure 1.
The analysis and prediction of civic complaints can contribute to better organization of cities and increase the quality of life for citizens.
For more information, contact dr. ing. Marwan Hassani.