The technical definition of pattern recognition is as follows: the calling out or identification (recognition) of patterns and regularities in data based on prior knowledge or the information extracted from the pattern.
Pattern recognition is considered by some to be a mature but still exciting and fast developing field, underpinning developments in cognitive fields such as computer vision, image processing, text, document analysis, and even neural networks.
The discipline of pattern recognition is continuing to find applications in fast-emerging areas every day.
Using data in this way is important, in that it provides insightful analytics to help align both recruitment efforts as well as coaching and team strategies aimed at the creation of successful—or winning—teams.
How will this work? Let's think about a similar problem.
We might look at a pattern recognition use case in the venue of e-commerce. E-commerce is both promising and challenging since e-commerce has been growing significantly, and yet competitive pressure is significant. Customers are mostly not loyal, arriving via search engines and online advertisements. On search engines, they are typically presented with multiple search results, including prices, which puts pressure on margins.
Typically, these organizations develop independent processes for single or directly related information, such as which products to stock, what prices to set, and what advertisement investments might be most efficient.
Recent blog posts and information online describe what might be a common sentiment on this type of process development, that various data sources and results yielded from analytical tools still require a human expert to intervene.
Returning to our use case, athletes can be evaluated on a number of data points and performance information, which may contribute to our successful position predictions, which will (hopefully) help our team's coaching staff make informed decisions and focus on the best target athletes when recruiting for a position or positions: pattern recruiting!
Actually, shifting our team from spreadsheet-based charting and graphing of athlete data to using Watson Analytics for analyzing the volumes of collected data that relate to each positional need is what this project is about: we will be mainly focusing on the following:
- Freeing the team's management from simplistic two-dimensional graphical analysis by providing multidimensional
- Providing high-quality and innovative designs that are easily interchangeable, so the visualizations have the utmost impact, making it easier to recruit, draft, and train athletes by position.