Towards a more efficient and personalised advertisement content in on-line social networks

Knowing what potential clients want, is the most important issue for companies. The current situation of social communication is generating a lot of information about users, such as favourite sites, food, politic tendencies, hopes or needs. This information has an incalculable value for marketing interests but, unfortunately, it is not trivial to process it. One way to obtain this information without disturbing the users is to store their searches, used words on the Internet and clicked items, to construct specific profiles. But, the problem with these techniques is that they do not retrieve current information about the targeted user because they gather information that may or may not be up to date. In light of this background, we propose a methodology to obtain up to date information about user's interests, likes and needs by analysing users conversations in social networks and instant messaging systems to generate personalised and interesting advertisement with better impact and higher success rates.