In my seminar, I explained trend detection in numbers and in text along with its visualization. I underlined few methods viz. Simple Moving Average(SMA) and Exponential Moving Average(EMA) to achieve trend detection in numbers and provided a prototype visualizing the same. I introduced you to the possible approaches to achieve trend detection in text. In the later part, I discussed Custom Search Application, Apache Solr and its features and provided a demonstration of its basic operations. Lastly, I briefly introduced the topics Semantic search and Linked data approach and its extension to Solr using Linked Media Framework.
Tag Archives: trend detection
The ThemeRiver™ visualization helps users identify time-related patterns, trends, and relationships across a large collection of documents. The themes in the collection are represented by a “river” that flows left to right through time. The river widens or narrows to depict changes in the collective strength of selected themes in the underlying documents. Individual themes are represented as colored “currents” flowing within the river. The theme currents narrow or widen to indicate changes in individual theme strength at any point in time.
An explanatory video:
- “ThemeRiver™*:: In Search of Trends, Patterns, and Relationships” and
- “ThemeRiver: Visualizing Theme Changes over Time”
are uploaded under pgpushpin’s documents at Mendeley.
This topic is divided into 3 parts viz.
1. Trend detection in numbers:
More: Moving average and its classification, predictive analysis and forecasts, visualization.
Example: Stock markets
Useful tool(s): MS Excel
2. Trend detection in text:
More: Term document matrix, comparisons, how can we achieve it, mathematical models, can we use Java/C#, visualizations, ThemeRiver.
3. Custom Search Applications:
More: Apache Solr, web services, semantic search, possible linked data extensions.
Currently I am reading whatever I come across about Trend detection. I am also learning various techniques with MS Excel. I have added a few papers those I found useful to Dropbox(links provided below).
Access e-books at Dropbox and papers at Mendeley in folder “Trend Detection” under PG PUSHPIN group.