Every year, thousands of overseas students come to New Zealand for their study. Due to language barriers or lack of familiarity with the local lifestyles, it is a challenging task for the students to find appropriate accommodation. Traditionally, the students find their accommodation out of a limited range of candidates by manually searching online advertisements published on local community websites. However, there is a high chance of developing a bad relationship between a student and an accommodation service provider after they live together in an apartment or house because of some subtle factors that are not explicitly described in the advertisements or discussed during the negotiation. Frequent relocation or bad relationships with the accommodation providers can pose severely negative effects on overseas students’ study and daily life. Although some popular platforms such as AirBnB provide accommodation seeking and booking services, they are mainly designed for tourists rather overseas students who usually need special consideration. This project proposes to implement a personalised accommodation recommender system for overseas students based on social media data analysis. The rationality of the proposal is that by analysing the social media data generated from overseas students and accommodation service providers, we can make use of the findings to optimise the matching between students and accommodation providers. The state-of-the-art social media mining techniques and recommendation algorithms will be leveraged to achieve highly personalised accommodation recommendation services. To ensure the scalability of the solutions, we also propose to implement the system based on cloud computing services. The main expected outcome of this project is an accommodation recommendation mobile app together with cloud-based back-end data collecting and data analysing servers. The students are expected to complete the following tasks: 1). Implementing the cloud-based architecture for social media data collecting and analysis; 2). Conducting experiments to compare the state-of-the-art recommendation algorithms, and selecting one or an ensemble of these methods for the project; 3). Developing the corresponding mobile app; 4). (Optionally) Proposing a novel recommendation algorithm to achieve better performance by applying certain machine learning techniques such as deep learning. Through this project, the students can expect to learn the following things: 1). Cloud-based programming paradigms, platforms and tools; 2). The state-of-the-art recommendation algorithms, and machine learning tools; 3). Social media data retrieval and mining techniques and tools; 4). Research and development skills; 5). (Potentially) Academic publications.
Undergraduate
The main expected outcome of this project is an accommodation recommendation mobile app together with cloud-based back-end data collecting and data analysing servers.
Through this project, the students can expect to learn the following things: 1). Cloud-based programming paradigms, platforms and tools; 2). The state-of-the-art recommendation algorithms, and machine learning tools; 3). Social media data retrieval and mining techniques and tools; 4). Research and development skills; 5). (Potentially) Academic publications.
None
Lab allocations have not been finalised