Big Data

The book​ is Out!

Posted by

1

The printed book is out!

Four years in the making, nothing on the standards of George R. R. Martin, but still a loong time. I was happy to see that last week, before being released fully, it appeared amongst the 10th most sold in Mannings early releases.
Thanks to all who bought it, supported it, reviewed it and waited so long for the final version!

screenshot 2019-01-18 15.24.06

I am delighted it is completed and I hope that you will enjoy it. Please feel free to comment, review or discuss with me. I also do talks if there is an audience that would like to hear about recommenders.

For now, happy days! Can’t wait to hold a paper copy later this week.

The printed book (and the ebook) are available here and will be for sale on all good webshops in the near future.

Get the book here

Introducing Practical Recommender Systems

Posted by

1

Practical Recommender Systems

Front page of Practical Recommender Systems

For a computer scientist like me, the world of IT is such an exciting place! Since I started at  university, I have seen the creation of companies like Amazon and Google, and later Netflix. They were for sure lucky to be in the right place at the right time. But it was ingenuity that has kept them in the market. What they did is a long story, but what I find interesting is that they have taken large quantities of content and made it accessible to the masses.

One of the advantages of being an internet business is the fact that you are not limited by physical walls like traditional shops and your list of products can be close to never ending. If a physical store was truly so vast, customers would struggle to find anything and  simply get lost. They would probably  go to the shop next door, which has fewer products and buy things that are not exactly what they wanted, but are easily accessible.

Offering lots of content does not ensure success, not even if you have precisely what your users want. Often 20% of your content will produce 80% of your business, if you can match the rest of the 80% of the content with your users, you will have more happy users and more business. The problem of activating the last 80% of the content is called the long tail problem.

A way to enhance the accessibility to the content for the users is to add a recommender system to you site. This can attempt to predict what your customers want and serve it to them.

Implementation of Recommender systems is an intriguing task. The actual algorithms like collaborative or content-based filtering are just a small part of it. If you do not feed the algorithm with the right data, it will not produce anything worth looking at. Using user ratings will often not produce the results that users want. Looking at context is also often something worth thinking about. And when it is all implemented and running, how do you know that it is working, how do you measure improvements?

I never found a book answering these questions; I found lots of good books explaining how to implement the algorithms mentioned above, but never a book that described everything around as well. So I started working on one. It just came out in an early release at Manning

Go and have a look, the first chapter is free!

Manning.com/falk.

Big Data, The silver bullet ?

Posted by

0

gaussiandistproblemA sign that Big Data is Big can be seen in the fact that the term Big Data has found its way all the way to the average Danish newspaper reader. Politiken, the biggest newspaper in Denmark, had it as a theme and wrote 6 pages about it. [among others this one 1] not long ago.

The newspaper article refers to hungry.dk – a Danish company, to illustrate the use of Big Data in a business.

hungry.dk provides a portal for take-away restaurants and is highlighted because they are using data from one of the new data interfaces delivered by the Danish state as part of the digitaliser.dk initiative. They retrieve smiley data for restaurants and merge it with their data on restaurants, enabling them to remove places that do not comply with the Danish food regulations. This, the article says, is one example which describes how companies can use Big Data, which is something that we are lacking behind in Denmark, and thereby loosing workplaces and competition edge, by not taking advantage of it [ 3 ].

I think its great that hungry.dk merges its list of takeout places with the smiley database, and its a great example of how businesses should take advantage of the data provided by the state. But for me Big Data is about analysing large dataset, to find patterns or to use it to predict the future, not so much about merging data from different sources. Merging data from different sources could very well be a step in a Big Data problem, but is not a Big Data problem in it self.

Either way, Big Data is something that is on a lot of peoples minds, and a tool that most companies should be consider using. The Data that the Danish state provides are there to be used, and even if you could ask why the state should pay for supporting high usages of the databases to enable businesses to bloom.

But where to start? Most books and websites makes it sounds like its about asking for the key at the Big Data engineer, and then the river of knowledge will magically start flowing out of your databases, and make exactly your company special. But is it that easy?

First point of order is the consider what data you got, or how to get it. Do you save backups? Ensure that all customer information is not a critical item, or actually save it in a way that makes it retrievable?. The collection of data can also be from an public API, or get data from Twitter to find trends and moods of things, or make a smart Facebook app so people completely voluntarily tells you all their secrets about everything from their dishwasher breaking to where they want to go for holiday next year ( imagine if a dishwasher seller could get info like that, or easyjet knew where people dreamt about going)(by the way are you sure that your facebook birthday app doesnt already collect data like that?)

When you got the data, and got enough to have a statistical relevancy (se 5) you can introduce the data to a statistician and her machine to enable them to learn something from the data. Actually you should probably include her when collecting the data too.

With success you can optimise your sale, your campaigns or who you build bridges better. There is virtually no limits to what you can make a machine help you with, if only you know how to teach it. An example where many businesses claims to gain on using machine learning is with recommendation systems, were it is said to add up to 10% more sale. [The classic examples netflix.com and Amazon.co.uk].

It sounds easy, but like most investigations, a single hitch can end in false conclusions, which will be hard to spot and resolve.

“The way to turn data into insight is to squash the notion that big data is a silver bullet. We preach that data and analytics is important but then we empower people to be curious and ask questions and get involved in big data analytics.” [ 6 ]

Everybody is Talking About Big Data.

Posted by

0

2014-06-13 06.17.18Everybody is talking about it, everybody is saying that they will soon have a version ready that will utilize the heaps of data, which are piling up in databases around us. But what is actually possible to achieve with it? Some say EVERYTHING, others are a bit more sceptical and think:

it’s being paraded around as a magic bullet, raising unrealistic expectations that will surely be disappointed. – Cathy O’Neil and Rachel Schutt in “Doing Data Science”

In my opinion, Big Data can be used for many things, but like everything using statistic, you should remember that correlation does not imply causation – just because something happens just after something else, it does not imply that one is a reaction to the other. Manipulated correctly data can prove almost any thesis, and its contradicting thesis. It is exciting to search for patterns or structure in the sea of data, to seek out information which no man has seen before, but be careful and sceptical always, especially when the results are too good.

I think its interesting that people can analyse data and find that children performs better in school when they eat breakfast every day[1], but personally I am more into predicting things whether it is to recommend good books, predict earthquakes or finding pregnant women from they shopping habits[2], is incredible cool.

I have been working with recommendation systems, studied Machine learning at the university and I am now working with it. I will always try to collect new ideas and learn more, which I intend to write about here. Many can be also found in Danish here at QED.dk

My hope is that this blog can be a place for people to come and read new interesting posts on Big Data and Machine Learning, but also please add to the discussion in comments or as guest bloggers.

Thank you for reading this, hope to see you again!