This post was originally published on Ascribe’s blog.
Like any other form of mining, data mining can be hard, dirty work if you don’t have the right tools.
Many customer experience professionals resort to hours of reading or excel manipulation to extract what they can out of their customers’ open-ended feedback. It might be better than not using it, but it is like using a rock hammer to dig for oil – you may never get there.
To transform your feedback from unstructured to structured for actionable insights, you need far more than a rock hammer. There are several kinds of “power tools” you can use for data mining depending on your comment volume, complexity, cost and maturity.
Here are a few examples:
Data Mining Technology Options
If you have large data sets and the need to derive meaning, develop taxonomy or access a query tool, rules-based text analytics (NLP) may be the right solution. This technology uses lexicons or dictionaries alongside series of deterministic rules to identify topics or sentiment, such as positivity or negativity.
If your program is more about large-scale, repetitive tasks, machine learning might be in order. Machine Learning uses artificial intelligence (AI) to learn how to categorize and interpret text automatically from a sample of manually classified training examples, so once trained, it can run as an automated process with minimal intervention.
If the highest degree of accuracy is required, you might do well with semi-automated classification. These methods organize the work intelligently, and optimize human decision-making in classifying customer comments by using powerful searches.
Selecting the Right Set of Tools
Sometimes, you need more than one tool to finish the job quickly and well. Your needs may change over time, or your data may become more or less complex. Be sure to consider all the possibilities as you select the technologies to use for your data mining and analysis efforts.