Paste your Google Webmaster Tools verification code here

Acte presentació llibre INTRODUCCIÓN A APACHE SPARK para programar el Big Data

Tenim el plaer d’informar-vos de la presentació del llibre ‘Introducción a Apache Spark’ dels autors Mario Macías, Mauro Gómez, Rubèn Tous i Jordi Torres. Es tracta del primer llibre sobre Apache Spark escrit en castellà amb el suport de MATEI ZAHARIA (creador d’Spark, CTO a Databricks i vicepresident d’Apache Spark) que n’ha escrit el pròleg.

Apache Spark és el software de càlcul distribuit (amb especial énfasi a aplicacions d’anàlisi de dades i machine learning) més popular actualment. Si hi teniu interès o el voleu conèixer, no dubteu a visitar el Barcelona Spark Meetup!

L’event serà el proper 3 de novembre a la UPC. Per assistir-hi, cal registrar-se a http://www.meetup.com/es/Spark-Barcelona/events/226143035/




Video and highlights of the event From Customer to Revenue

From Customer to Revenue, the theme of our third event, brought together two speakers from industry to address challenges in revenue management.

We started the evening with a series of updates: 1) The BCNanalytics job board is a success and can be found on our homepage. 2) On 14th and 15th of November we will organize a Hackathon in collaboration with Social Point, further details can be found here. 3) We supported an intern named Enrico Kunz in analyzing twitter feeds to identify deals from tweets. Finally, our two industry speakers,  Sharon Biggar from Social Point and Francesc Paloma from Vueling, took the stage to share their experiences in revenue management.

Sharon from Social Point

Sharon started by explaining Social Point’s business model of free to play and how their games should not be seen as finalized products, but rather their games are continuously updated with content and feature updates. She went over various examples of how analytics helped Social Point identify points of improvement in their games. Analysing at which point in the gaming experience users drop out. Examining correlations between high transactions and high use of game features. Understanding the best use of bundled offers of in-game purchases to help convince users to make their first transaction. A general good rule of thumb that has shown to work well in Social Point’s free-to-play approach is “If the game is fun, the money will come”.

Francesc from Vueling

Francesc explained how Vueling has evolved in being data-driven since 2007. At Vueling it is important to identify periods that will result in a high number of bookings early on and have ticket prices that correspond to high demand. This means accounting for the different holiday periods in different countries, but also special events such as the Mobile World Congress in Barcelona. In recent years, however, revenue management for airlines has mainly developed in the area of ancillary revenues, or revenues from non-ticket sources, such as seat reservation or on-board food services. Francesc gave several examples on how data helped in deciding on the pricing strategy such as fixing a ticket price for 24 hours before booking or using a dynamic price system for seat reservations.

Sharon and Francesc

After the presentations the audience got the opportunity to ask questions, allowing the speakers to address some further details of their approaches. The session closed with drinks and a chance for visitors to do some networking and discuss the topics presented.




Barcelona Gaming Data Hackathon. Ready?

 

Logo SocialPoint

BCNAnalytics

 

 

 

BCNAnalytics and SocialPoint are pleased to invite you to the first Barcelona Gaming Data Hackathon.  40 participants. 10 teams. +€1.5k euros in prizes. 24 hours. Food, t-shirts, great views and awesome people. R, Python, ML, … Ready?

The competition will focus on analyzing the data about the Dragon City Game (iOS and Android). Teams will participate in 2 different tracks: the accuracy track about churn detection and the business insights track, about finding actionable insights from the given datasets.

Register!

Social Point Offices

 


Competitions

The hackathon has two tracks to compete for. They are both about the Dragon City (iOS and Android).

  • Accuracy track: given data about how online players behave in Dragon City game, you will need to predict when players are likely to abandon the game using data from the players first 48hrs in the game (churn model).  As in Kaggle competitions,  teams will have 3 data sets: the training set, the test set1 and the test set2.  The training set  will have feature variables and a target variable to train the models. Once a model is trained, you can use test set (provided without the target variable) to submit your results and get instant feedback of the performance of the model.  Whenever you feel confident with your result, you will have to use test2 to make the final submission. Only results from test set2 will be used to determine the winners of this track. Each team will have 24 hours to build and submit their models (from Saturday 14th November to Sunday 15th November).
  • Business Insights track: participants will need to present actionable insights from the dataset provided (e.g. increasing the difficulty in level 12 will decrease 13% the drop out on that level). Each team will have 8 minutes to expose the finded results to the jury.  The results will be evaluated based on the impact of the insights, the actionability, the understandability of its communication and the novelty of the results. Each team will have one hour to prepare the presentation after the accuracy competition closes, although we encourage to be thinking about those insights while analyzing the data for the Accuracy track.

Prizes

  • Accuracy track: 1st prize 750€, 2nd prize 500€
  • Business insights track: unique prize 500€

Schedule

Saturday 14th November

  • 9:00 Opening and participants check in.
  • 9:30 Presentation by BCNAnalytics and SocialPoint. Team Formation
  • 10:00 Competition start. Data sets are released and submission platform begins accepting submissions.
  • 14:00 Lunch all together to a cool place :)
  • 20:00 SocialPoint offices closes

Sunday 15th November

  • 9:00 SocialPoint offices reopen
  • 10:00 Accuracy competition closes. Submission platform stops accepting submissions. Teams can prepare the presentation for the business track.
  • 11:00 Begin presentations
  • 13:00 Jury delivery
  • 13:30 Prizes
  • 14:30 Closing and optional lunch.

Participants and registration process

40 participants, 4 members per team.

To participate to the hackathon, start registering here, providing your Linkedin and Github profiles. If you already have a team formed, indicated the name of it in the registration form. If you don’t have a team, register and we will assign you a team.

The deadline for registering is Monday 9th November. The acceptance process is based on the public profiles provided. Once you or your team has been accepted, you will be notified about it.


Rules

  • Each participant can only participate in one team.
  • In order to opt to prizes, all teams must participate in both competitions.

This competition would not be possible without the help of SocialPoint. Thanks!

Logo SocialPoint

 

 




Upcoming event: From Customer to Revenue

We are very pleased to announce an upcoming event at BCNAnalytics. In this occasion we will be honoured with Sharon Biggar, head of Analytics at Social Point and Francesc Paloma, head of revenue management at Vueling. The format will be similar to the previous events: small presentations of the speakers, Q&A and finally networking time.

When: 21st September, at 19:00 (doors open at 18:30)

Where: Mobile World Center (next to Plaça Catalunya).
Fontanella, 2. Barcelona.

You can register here!

sharon-headSharon Biggar is the Head of Analytics at Social Point, one of the world’s leading developers of games for mobile, with more than 50 million active monthly players generating terabytes of data daily. There, she ensures that SP has the Big Data architecture ready to support the user base, analyses game play and user comments to improve the games and analyses user behavior to help customize user experience with predictive models.

Prior to that, Sharon was the CEO of Path Intelligence, a technology that provides online analytics to the offline world, analyzing customer behavior in shopping malls, retailers and entertainment arenas.

Sharon is a kiwi (from New Zeland). She holds a degree in Music and a degree in Economics, from the University of Auckland, and an MBA from the MIT  – Sloan School of Management.

Francesc PalomaFrancesc Paloma is the Director of Revenue Management & Pricing  at Vueling, where he has held various positions related to revenue management and pricing. He is responsible for dynamic flight price setting and other revenue management initiatives. Revenue management has become a more sophisticated analysis in recent years as competition has increased and data is more readily available. Vueling has also been innovative in terms of revenue strategies.

Prior to that, Francesc has lead analytics projects applied to social exclusion in Barcelona and was an auditor of lotteries at Loto Catalunya. He has also been a professor of Mathematics at the School of Engineering at UPC.

Francesc hold a degree in Math from UPC and an MBA from UPF, as well as a PDD from IESE.



Nonregular event: Data Science at Booking

We are pleased to announce the following event, sponsored by Booking.com and in collaboration with Gemleb. In this session, people from Booking.com will explain us what kind of problems they deal with related to data science and machine learning (the talks may be more technical oriented than usual).

1) Data Science: For Fun and For Profit

Data Science is relatively new, but the ideas and techniques that form the underpinnings for this evidence-oriented discipline have a solid foundation in hundreds of years of scientific development. In order to understand the new science of data, one must first understand the science of science.
The Scientific Method, the unintended effects of repeated significance testing and Simpson’s paradox: this talk will focus on the practical applications of the theoretical constructs that lie at the heart of Data Science; and expand on some potential pitfalls of statistical analysis that you are likely to encounter when venturing into the field.

Bio: Lukas Vermeer (Data Scientist, Booking.com@lukasvermeer)

Lukas is an experienced data science professional with a background in computing science and online machine learning for real time decision support. A strong advocate of “Evidence-Based Everything”, he is forever learning and helping machines do the same. As a Data Scientist at Booking.com, the world’s leading accommodation website, Lukas is exploring novel ways to make booking hotels online into a more personal experience.

2) Topic Modelling on Travel Data.

Booking.com collects millions of diverse endorsements from its users, for example, London endorsed for Shopping, Brussels for Chocolate, Athens for Museums and Barcelona for its nightlife. These endorsements are organised using Latent Dirichlet Allocation to a set of topics and used to personalise the Email-Marketing campaign of Booking.com. The results from experiments on more than 40 million unique users demonstrate the conceptual value of the discovered topics.

Bio: Athanasios Noulas

Athanasios Noulas completed his PhD in Machine Learning in the University of Amsterdam where he focused on Dynamic Bayesian Networks and Deep Learning. He then worked as a strategist in Source Capital where he developed algorithms for high frequency automated training. He is currently working for Booking.com as a Data Scientist in the Visitor Profiling team, where he performs analysis on user-behaviour and implements algorithms that adjust the web-site to the user’s needs.

At the end we will have some drinks and time for networking.

If you want to attend, please register via meetup.

rgb_std_logo_8_1_




Idescat’s 250m grid population data

We are glad to announce that population data in a 250m grid from Idescat is now available. Here you can find the following fields from the 2011 census ‘Registre de Població’ from January 1st of 2014:

  • Total population
  • Men and women population
  • Population by ages: 0-14, 15-64, more than 64 and more than 85


Click to enlarge
Click to enlarge

You can also find the (R) code to create the map of any of the variables above in the BcnAnalytics github account

We would like to thank Idescat for their support.




Personalization in CRM event review

On March 9th, 2015 BCNanalytics was proud to host its second conference, “Personalisation in CRM”.

  • 198 people registered
  • 165 people atteneded
  • 83 empty beer bottles by the end of the event

The event was hosted at the Mobile World Centre and was attended by people representing  industries like banking, e-commerce, consulting, gaming, fashion, telecom, as well as academia and independent entrepreneurs.

To start off the event we had Manuel Bruscas, co-founder of BCNanalytics, present some of the key news of our organisation, like the creation of a job board on our site and some of the potential events we would like to organise like a hackathon or workshops.

Manuel presenting

We had Pau Agulló, also co-founder of BCNanalytics, introduce our two keynote speakers and later serve as moderator during the Q&A session.

Pau  presenting speakers

Our first speaker was Pier Paolo Rossi, current director of Business Intelligence at Banc Sabadell. Pier Paolo’s presentation covered a wide range of examples coming from his multi-industry experience having worked in telecomms (Movistar), retail (Caprabo) and lately banking (Banc Sabadell). He showed how personalisation can be applied to pricing, product, information layout and other areas of your business to expand the lifetime engagement and experience of your customers. His talk highlighted two points we would all do well to remember: “you can’t manage what you can’t measure” and “your customer is the most important thing”.

Pier Paolo Rossi

Our second speaker was Llorenç Solà, Director of CRM and Customer Strategy for Vistaprint Europe. Llorenç explained us that using personal content in communication wins over non-personalised approaches because it allows a business or organization to become more relevant to their users. He also reminded us that appropriate personalisation is by no means easy to achieve, since it always involves: data storage, analysis, statistical modelling and a user friendly strategy around frequency of touch points. There was particular emphasis given on the last point applied to email practices. In his opinion if long term performance is desired then email campaigns should not be measured simply by gross revenue. That would only lead to the controversial practice of mass emailing, which can arguably degrade your ability to communicate and influence your customers. Instead performance should be measured by the efficiency by which you get customers to interact with your communication.

Llorenç Solà

After the presentations our keynote speakers formed a panel and we discussed what had been presented. The audience was actively engaged with many questions for our presentors. Regarding the role of privacy in personalisation, one opinion was that if done right, personalisation is welcomed by customers. When asked which channel to use for communication the answer was that you should make your customer choose the channel for communication that they want. Other questions addressed the role of people and technology, as well as further discussion on the impact of limiting the amount of emails sent to customers in favor of more personalised content.

The conference closed with a networking event with free beers provided by our sponser Estrella Damm. People took this opportunity to ask further personal questions to our keynote speakers as well as meet fellow members of Barcelona’s analytics community. Thank you all for making this event a success. Please help us decide on our next event as well as ideas for other events by filling out this survey.

Beer time!!

 




Upcoming event: 9th March Personalisation applied to CRM

We are pleased to announce our upcoming BcnAnalytics meeting. The focus will be on: “Personalisation applied to CRM”. In other words we address one of the key topics for many organisations: ‘how to better manage the relationship with your customers’.We will have a panel with Pier Paolo Rossi (Director Inteligencia y Desarrollo de Negocio at Banco Sabadell), Llorenç Sola (CRM Director – Vistaprint Europe) and Inés Urés (CMO at Groupalia).

If you want to attend, please register at the following link.

When:

9th March, at 19:00 (doors open at 18:30)

Where:

Mobile World Center (next to Plaça Catalunya).

Fontanella, 2. Barcelona.

Who:

Pier Paolo RossiPier Paolo Rossi  is the Business Intelligence Director at Banc Sabadell since September 2014. Previous to that, he was the Customer Marketing Director at Caprabo. He has a broad range of experience applying data analysis in such different industries as retail and banking . He holds an MBA from IE and a degree in Industrial Engineering from UPC with the specialization in nuclear energy.

 

Llorenç SolàLlorenç Solà  is Head of CRM and Customer Strategy at Vistaprint Europe (Nasdaq: VPRT). More than 15 years developing data driven marketing strategies and CRM. International career in strategy consulting advising multinationals in Europe, America and Middle East. He is also one of the cofounders of BCNAnalytics.

 

inesInés Ures is the CMO at Groupalia, e-commerce leader in health & beauty, leisure and retail discounts. She’s a mathematician and an engineer, and a lover of data based marketing. In the past, Inés has held different positions in marketing and project management in Atrapalo.com and she has also worked at Deloitte and Sara Lee.

 




Barcelona Data Lovers