comment from post 1
Most qualitative data is “Unstructured Data” which can come in the form of documents, images, audio, and video.
The easiest things researchers can do to make Unstructured Data more useful for analysis is:
1.Make it easily searchable
2.Convert it into a structured schema, which can then be analyzed using quantitative techniques
On the first point, researchers can feed documents to full text search engines which allow the data to be easily retrieved. They can also configure full text search engines to perform faceted search allowing to attach Metadata facets (e.g. Author, Media Type, Creation Date, etc.) allowing to easily look up support qualitative data to support their quantitative analysis. The same search engine can also support the indexing of non-text media (e.g. Video) by indexing the documents Metadata/contextual attributes.
On the second point, there are numerous ways of deriving Structured Data (which can be quantitatively analyzed) from qualitative Unstructured Data. But it all depends on what and how you derive the Structured Data. For example, you can derive n-grams (continuous sequences of n [e.g. 3 or 4] words, and then analyze those n-grams to see what the most popular terms are within a subset of documents (or for example Tweets) in order to get an idea of what topics are most discussed or what phrasings are popular – the zeitgeist. Additionally, algorithms like Word2Vec and Doc2Vec can be used to quantify sentiment of a document (or Tweet). For analyzing video you may want to have a person manually transcribe all consumer mentions of a product. There are Machine Learning algorithms that can already transcribe voice and recognize. Machine Learning and Deep Learning applications which can derive useful and accurate quantitative data from qualitative information will play a huge role in the future of Analytics. But manual methods like using the Amazon Mechanical Turk, or a combination of both, are also effective approaches to deriving Quantitative Structured Data from Qualitative Unstructured Data.
I NEED YOU TO COMMENT FROM THIS POST, 150 WORDS NEEDED AND A REFERENCE PLEASE