One of the ways we try to take advantage of technology to improve our investment process is with machine learning algorithms.  In simple terms, we gather large amounts of historical data and using that data we “teach” a computer system to distinguish between good investments and weaker ones by finding relationships between the historical data and subsequent investment performance. Our experience with these systems leads us to think that they have a useful role to play in sorting the wheat from the chaff. We also find that this approach has particular power at the smaller end of the market, where lack of analyst attention can more easily result in interesting opportunities going unnoticed for a while. With this in mind, we have run our Australian machine learning model over the $50-$300m market cap section of the ASX. The results are below, read the full article for more detail: (VIEW LINK)