Bachelor Project: Alignment of movie databases

Today, the internet is the largest distributed database. When data from two sources is integrated into a single database one major topic is to detect records that correspond to the same entity. One typical scenario for this tasks are price comparison systems. Here, offers from different shops are automatically crawled and you have to detect which offers correspond to the same product. The task of solving such linkage problems is known among others as record linkage, duplicate detection or object identification. In this bachelor project the simplified problem of aligning two movie databases should be addressed.

The following tasks should be investigated:

• Analysing the data.
• Create rules/ methods for aligning two movies. Optionally, the method can be learned by machine learning techniques.
• Implement an algorithm for the task.
• Evaluate the quality of the predictions.