Data mining, inbound marketing and twitter hashtags analysis


The Inbound marketing is a methodology that covers a broad number of techniques for on-line marketing, including blogging , social media and many others ,  Scraping data from web is the first step for creating clustering and data mining models to better understand your data and make them more useful and easier to use to improve your brand.

Besides using these techniques to sell more, they can encourage collaboration between community members. This is the case of a new platform, Greenius http://greeni.us/, that will be available soon, and  will put in contact   people all over the world interested in gardening and farming ,both in the countryside or in urban areas.  Every user of this platform would share information about their vegetable garden, their allotment or what they are growing in their balcony. The community in this platform will probably find interesting the suggestions  given by the application.

Advanced data mining tasks will be used to dynamically provide the platform with this content. Specifically, the platform will use association rule mining in twitter-like hashtags in order to find the likelihood of their  co-occurrence.  This technique, widely used in the basket  market analysis, where customer shopping data  can show what articles are purchased together,  can be translated to hashtag analysis after some processing to preserve the changing dynamic of the system. The extracted rules can be considered as events happening at different points in time and here is where temporal data analysis takes place and rules analyzed  with this new dimension: the time. The recommendations need to evolve to the extent the system evolves. The potential of this type of systems can be easily understood in many real systems in which the trend changes can be discovered in tweets and decisions corrected on the fly.

+ No hay comentarios

Añade el tuyo