Question # 1: Is ‘targeted’ advertising ‘smart’ advertising?

Answer: Indeed, incredibly smart. Apart from the fact that it is ‘targeted’ as it generates relevant advertisements triggered by the user’s queries here are a few more reasons.

• Aware of the user’s geographic location through the IP address and can micro-target an audience pretty much to a radius of a few kilometers.
• Based on the demographics – Gender, Age, Status etc.,
• Recognize the device from which the query is being fired and return ads accordingly,
• Using the data from services like personal search or clicks and predict the behavior of an individual and serve ads accordingly
• Using data from your social graph for example, if you ‘liked’ a product on facebook or the type of places you’ve checked in on ‘foursquare’ etc.,

Question # 2: Can ‘targeted’ advertising grow even smarter?

Answer: Again, indeed. Consider the following scenarios.

I’ve been observing a deluge of Chrome ads these days …both as video embedded text ads and single block text ads on ‘Youtube’. I like chrome and it is my favorite browser but is Google smart enough not to push these ads onto somebody already using chrome? No, at least not yet. With a growing population of people online using chrome this practice, if not corrected, will only prove increasingly fruitless. This stems a new question. Why can’t Google detect the browser information and trigger ads accordingly? I can imagine a world of opportunities if this happens.  Different browsers have different capabilities and can run applications or services that are aligned to those capabilities etc.,

I’ve also been observing another set of ads that Google has been presenting. The query site:[DOMAIN NAME.COM] triggers an ad for  ‘Google Webmaster Tools’.  But the ad gets triggered only for some domain names.  Quick research seems to reveal that these ads don’t appear for the supposedly chosen few ‘white hat’ or well know websites (though I could be quickly rebuked on that), which is fine. For the rest, the problem is that Google doesn’t seem to be checking whether the domain name used in the query already is using the advertised application or not, in this case Google Webmaster Tools.  Again, this is targeted advertising but definitely not smart enough. Why waste so many impressions when you can save on them by putting a check in place to confirm whether the user is already using Google Webmaster Tools, especially if the user is making the queries logged in ?

Now, I do understand that often these changes are not as easy as I’m proposing them and that there could either be a ton of technical constraints that are acting as bottlenecks or simply the effort v/s outcome tradeoff is not worth it, at least not yet.  But I see incredible potential here especially as the idea of OpenSocial and OAuth become more popular and with so many companies opening up their hoods through the APIs, advertising for these very applications can make use of this development and gain an understanding of what applications a user has already been using and present ads in a more smarter way.

So, according to ‘them’ the global meltdown fuelling the use of this paradigm is just no more than a coincidence. There is a ton of research being done in various fields like Technology Adoption, Information Security and Organizational Behavior. Like any other innovation there are early adopters of technology and later adopters of the technology. The early adopters are usually characterized by their risk-seeking or entrepreneurial mindsets and late adopters by their risk-averse mindsets. What is interesting is to see if the above mentioned mindset, let’s call it ‘Perceived Risk’ linearly reduces with time or not and what factors mediate this relationship. This definitely will have implications on the success or failure of an innovation and can be useful to innovators and technology adopters alike. There is an awesome video (2 parts) from a CloudCamp session in Frankfurt where Simon Wardley from Canonical, Sponsors for Ubuntu talks about the state of confusion when it comes to understanding what Cloud Computing really is. He describes how this paradigm of computing wasn’t realized overnight, how it traveled through the different phases that any innovation usually does and why its’ time really has come. Totally recommend it.

Initially there was a lot of confusion about the difference between the fields of Inferential Statistics and Data Mining and researchers around the world immediately started working on producing tons of literature to address this issue. Though there is no single universally accepted definition for either of them, many researchers have defined these fields through their individual application perspectives. From what I’ve gathered and known about these fields, this is how I draw a line between them.

Difference between Data mining & Inferential Statistics:

• Data Mining is a field where you discover hidden patterns from already existing large data sets. These hidden patterns discovered are later used for analysis and decision making scenarios in the area of concern. The process of using the discovered hidden patterns is also called as ‘Knowledge Discovery’. Inferential Statistics is the field where you prove or refute a pre-conceived hypothesis (or a null hypothesis) by performing classical statistical methods on a sample of a given population size.
• Data mining starts at an already existing database (usually large datasets) and Inferential Statistics generates its own database using sampling methods on data set.
• Data mining methods employed like classification, clustering, etc., scan the entire dataset in search of hidden patterns while classical statistical methods are run over only a small section of the dataset (the sample).

The above point also infers that Data mining methods are more computer intensive as they have to run through large data sets and hence should be used only when really needed.

So when do you use data mining over inferential statistics?  Well the answer is simple. If you don’t know what you are looking but want to make the best sense of the data you have then use data mining. If you know what you are looking for and want to back it up with proof by checking the data you have you should use inferential statistics.

Indian celebrities on twitter:

Sashi Tharoor

Narendra Modi

Barkha Dutt

Meghna Gulzar

Gul Panag