Atime-series database stores sequences of values or events obtained over repeated measurements of time (e.g., hourly, daily, weekly). Examples include data collected from the stock exchange, inventory control, and the observation of natural phenomena (like temperature and wind).
Hi Satnam,
I'm not sure if I understood your question correctly, but let m try to help you.
Basically, a clustering algorithm seeks for instances with similarities and create groups from them. They are used, among other things, to discover groups in your data that you don't know yet.
Association algorithms intend to find association between occurrences, for instance, if I have temperature '>40', wind is 'high'.
Clustering and association are unsupervised tasks, in other words, we don't have a label for each example.
On the other hand, a time series analysis algorithm may look for trends and are used in most cases for forecasting.
I hope that this answer can be useful for you.
Regards,
Fábio
Dear Mr. Satnam Singh, in the case of the chronological series, we intend to look for a seasonal trend focused on a study of the behavior of the equidistant time func tion (I recommend you see the studies presented by Box, G. and Jenkins, G. 1970) Time Series Analisys forecasting and control Available at: http://www.garfield.library.upenn.edu/classics1989/A1989AV48600001.pdf ), so you can better understand these aspects.
In the other case of using clustering techniques are generally to be able to group and characterize databases limited to specific periods or times of time and are not highly recommended in the medium and long term inference of the data ...
Hi all please specify your answer in 2 or 3 lines ..How we can differentiate between these algorithms..how a layman can understand basic difference between these algorithms
- Badges
- Science topic
- Similar topics
- Computer Science
- Data Mining