Skip to content. | Skip to navigation

Informatik 5
Information Systems
Prof. Dr. M. Jarke
Personal tools
You are here: Home Theses A Rule Based Recommendation System for Data Trend Visualisation


Prof. Dr. M. Jarke
RWTH Aachen
Informatik 5
Ahornstr. 55
D-52056 Aachen
Tel +49/241/8021501
Fax +49/241/8022321

How to find us

Annual Reports





A Rule Based Recommendation System for Data Trend Visualisation

Thesis type
  • Master
Status Running

Recommendation systems help users make decisions. With the advent of large, high-dimensional datasets there is a need for tools that can support rapid visual analysis. For a data analysts or a normal business user, it is a difficult task to select the related visualisation for a given dataset which is also the most appropriate one. Thus the main aim of this thesis work is to build a rule engine for time series data which will be able to collect the context of the data and can recommend visualisations accordingly.  The output of this thesis work will be used to build a visualization recommender system for mining the data for interesting values, trends, and patterns to speed up data analysis

Time-series dataset shows the time trend of the data. For the time series data, there are different forecasting methods to get the data trend for a particular dataset. Thus the proposed recommendation engine will recommend the corresponding charts by taking the data trend into account. The recommendation system should be based on the knowledge-based recommendation technique. Since we are not considering the user history part to the visualization, the other recommendation technique like collaborative filtering, content-based filtering or hybrid recommendation is not appropriate to the rule engine. In this way, we eliminate the drawback of having a "cold start" problem where the user history is very important to the system.

Magor Goals of this Thesis will be as follows: 

1. Discover Data Trend: One of the main objectives of this thesis work is to discover the trend of time-series dataset. 

2. Implementation of Rule Engine: The rule engine will be used for finding the related visualization technique. At first, the rule engine will construct a visualization model by transforming the previous step domain knowledge result.

3. Automated Data Visualization: Input data collection will be mapped to the prediction results to develop the final recommended visualization model. The system will visualize the charts automatically for the input dataset and recommend it to the end user

5. Selection of appropriate Dataset for evaluation of the rule engine

Below is a rough overview of the proposed model 


Recommendation system for data trend visualization




Document Actions