Summer Update

activity update

We just finished the most amazing quarter since our creation. We moved to DigitYser, found 2 new supporters with Sofina and the Brussels Region, we launched the second coding camp and datascience bootcamp. Unfortunately we also had to say goodbye to a good friend.

Moved to DigitYser & new core team.

DigitYser is the digital flagship of Brussels where communities gather to learn about, co-create and implement digital solutions focused on Blockchain, Data & AI , IOT and VR.

www.digityser.org

2nd Coding Camp and Data Science Bootcamp.

The candidate data scientist will sharpen his coding skills during the summer coding camp.

Companies will have the possibility to meet the candidates in person and offer them a job.

di-academy.com/bootcamp

Job Fair to meet the candidates of the Data Science Bootcamp.

Introducing our Summer Coding Camp participants to organisations interested to sponsor their Data Science Bootcamp is key to DigitYser’s mindset

Recruiting Interns.

We are recruiting interns and want to help them to get a much better insight into the following passions : AI – VR – Data – Blockchain – IoT – Social Media – Video.

Ending with sad news:

We end this overview with sad news as our friend Colin Molter has flown away. Colin was a respected data science professional & entrepreneur and Chief Data Science @ AXA.

You can leave a kind word for his family in the comments of the linkedin post.

Up to you ! Please help us.

  • Please share the info about the data science bootcamp to your colleagues.
  • Share the info of the job fair with your HR department.
  • Come and visit us at DigitYser and attend one of our events.
  • Write a nice message for Colin on our linkedin post.

Contact details & interesting links:

Join the Data Science Job Fair

Introducing our Summer Coding Camp participants to organisations interested to sponsor their Data Science Bootcamp is key to DigitYser’s mindset

jobfair4

We are so excited to have been contacted by so many talented professionals interested to join our Summer Coding Camp! This is a clear sign on the awareness towards the importance of developing and promoting Data Science skills.

This is a great prelude for the first Data Science Job Fair 2017 edition, where we bring in contact the participants of our Summer Coding Camp with organisations who are interested to sponsor their Data Science Bootcamp and to invite them to work for them. This is in line DigitYser’s mindset, which is highly focused on bringing together different stakeholders around the activities we organise.

A year ago, during each of our Data Science Job Fairs, an average of 20 companies (Innovation and Training Partners) were able to meet 40 participants of our Summer Coding Camp and Data ScienceBootcamp. This is a great opportunity which goes beyond a typical job fair, as the event’s format to both the training’s participants and the organisations to briefly present themselves to everyone, ensuring a clear visibility of their profiles and the added value they bring along.

The event is open to everyone: feel free to book your place and join us on August 1st. It is free of charge for the Summer Coding Camp participants and for our partners. Don’t hesitate to contact Philippe (pvanimpe@gmail.com) if you want to know all the benefits of being a partner.

If you want to join the Summer Coding Camp, you are still on time! Join our informative session on July 17th. or contact us asap via e-mail (training@di-academy.com).

Keep an eye on the upcoming events on our Facebook page, and join us at our Data Science events at DigitYser! We look forward to meeting you soon!

Announcing the second Data Science Bootcamp

bootcamp2

Why investing on a Data Science bootcamp is a win-win for both organisations and employees

“Once upon a time, when data was part of an isolated silo…”

The previous phrase perfectly reflects how data projects were perceived in the past. Long gone are those days! Nowadays, data is used in more integrated manner, allowing to connect the dots among different data pools and to predict trends. This digital transformation supports companies to be more proactive and remain competitive. In an era where terms like “Big Data” and “Blockchain” are ubiquitous, organisations need to ensure that the way they manage data is up to speed to take the full advantage of it.

A few weeks ago, during a panel discussion on the CDO role scope, Jonas Vandenbruaene pointed out that 40% of the cases when data projects do not meet expectations are related to lack of skills. In order to sort out this challenge, organisations need to actively support their employees to acquire and master data related skills. Furthermore, this requires to -on top of acquiring technical skills- understand how to correlate data to business decisions and processes.

The acquisition of technical skills for handling needs to be complemented with business acumen acquired through real life business cases

Hence, it is crucial that companies proactively encourage and support their employees to update their skill-set with a pragmatic approach. This certainly represents a win-win: while the organisation acquires state-of-the-art tools, the employees performance and motivation remain high, leading to a good employee retention level.

In order to help organisations to achieve this, at di-Academy we are thrilled to present the second edition of our Data Science bootcampThe program will take place at our brand new location (at DigitYser, centrally located next to metro Yser in Brussels) and is set up in two pillars:

  • The first pillar is a 7 week long Summer Camp (starting on August 7th.), providing knowledge on coding and technical tools via coached MOOC’s. Participants can also earn a certification on SAS, Microsoft Azure, SQL, Python, R, and Statistics.
  • The second pillar is a 12 week long Data Science Bootcamp (starting on October 2nd.), during which real business cases are presented by our partners on a basis of 2 days a week.

This scheme allows corporations two enrich their human talent in 2 directions:

  • Training their existing employees, allowing them to combine the bootcamp with their current job.
  • Meet highly skilled jobseekers who can be hired after the bootcamp. For this purpose, throughout the program, there will be several job fairs showcasing the participants’ talent.

Do you want to learn more details on these training and the job fairs? We invite you to join our presentation and Q&A session on July 3rd. and to contact us (training@di-academy.com, +32 (0) 470 365 507), so we can provide more details.

It is a pleasure for di-Academy to help talented companies and their employees developing digital skills. We look forward to be in touch with you and to keep alive the conversation on acquiring Data Science skills!

Press Release: Introducing the Data Innovation Summit 2017

On Thursday, March 30th, 2017, the 3rd edition of the Annual Data Innovation Summit will take place in Brussels, at the ING Marnix Building. This event, organized by the European Data Innovation Hub, is the #1 Data networking event in Belgium, bringing together the academia, public and private sector to discuss the latest trends on big data. This will be the meeting point for 500 data lovers to learn from top-nodge presentations, workshops and trainings.

During the event, 45 recognized speakers will share their experience related to the edition’s theme “Using data to build a better world”, reflected in the hashtag #Data4Good.  Among them there will be Bianca Debaets (State Secretary of the Brussels-Capital Region), Elena Bonfliglioli (Senior Director Health Industry EMEA at Microsoft), Kirk Borne (Principal Data Scientist at Booz Allen) and Stephen Brobst (CTO at Teradata) talking about how data, analytics and digitalization can be brought together to make sense in society.

From discussing the realistic perspectives of artificial intelligence to sharing best practices on using analytics to spot cybercriminals, the topics’ spectrum will cover artificial intelligence, machine learning, data driven marketing, social media, health and work performance improvement and many other topics reflecting the outstanding potential offered by data. There will be 10 in depth workshops taking place (5 in the morning and 5 in the afternoon) on digitalization and analytics platforms and software.

Additionally, the winners of the Dengue Hackathon (held on November 2016) will be on stage to receive their prize and share how open data can be used in the fight against dengue.

Hence, as Philippe Van Impe, Founder of the European Data Innovation Hub says:

“This event, and the enthusiasm of the lively community and volunteers supporting it, is a sign on how Belgium is becoming a key player in the European Data Science scene.”

The organizing team looks forward to welcoming the attendants.

For further contact, please refer to the hashtag #DISUMMIT and to the following contact details:

Name of Media Contact: Philippe Van Impe
Title of Media Contact: CEO and Founder
Company Name: European Data Innovation Hub
Contact Phone: 0477 237 842
Contact e-mail: pvanimpe@dihub.eu
Website URL : http://www.disummit.com

#disummit – Executive Workshops – Brussels 30 March – Register Now

Executive WorkshopsDuring this year’s summit you will hear stories on how data is used to build a better world.

On March 30th we have invited an impressive lineup of top speakers and prepared in depth training classes for managers and data scientists.

We are also offering you 5 exclusive C-Level workshops:

  • 29 March 16:00 – 3h Workshop from Dirk BorneCommunicating Data Literacy and the Value of Data to Clients and Colleagues
  • 30 March 10:45 – 2h Workshop from Geert VerstraetenIntroducing Predictive Analytics – more
  • 30 March 13:45 – 1h Workshop from Natalino BusaPositioning Open Source in your existing software architecture
  • 30 March 13:45 – 1h Workshop from Stephen BrobstLeadership and Digital Transformation – more
  • 30 March 17:00 – Closing Keynote from Kirk BorneA Data-rich World for a Better World: From Sensors to Sense-Making
  • Please check disummit.com form more exciting presentations.

Sept 8th 18:00 – Start of Data Science Antwerp

  • Prof David Martens, Vincent Van Look & Michaël Peeters are initiating this new meetup location for the data science community at the University of Antwerp.
    Bart De Wever will do the opening speech.Agenda :18.00 – 18.15  Introductie van Antwerp DSC18.15 – 18.45  Bart De Wever (mayor Antwerpen)

    18.45 – 19.15  TextGain: Real-time text mining applications: finding customers, cyberbullies and jihadist  by Guy De Pauw, Tom De Smedt and Walter Daelemans

    19.15 – 19.45  City of Things: Data Science with Internet of Things Data by Philip Leroux

    19.45 – 20.15  Telenet: Deriving insights for improved viewing experience through audience analytics by Kurt Vanmechelen

    20.15- 23.00   Networking

    Register her for this meetup:

    Each month we organize a Meetup in Brussels focused on a specific DataScience topic.

     

Summer Data Science activities in Belgium.

summer edition

We wish you happy holidays, in case you get bored check out our educational channel on youtube.

The European Data Innovation Hub is active during the summer.
Here is a short update from what to expect in the coming weeks:

Thank you for supporting the European Data Innovation Hub, we had a great academic year.

Philippe Van Impe
@pvanimpe
www.di-academy.com

Please forward the information about the data science bootcamp to your peers and friends.

How Tom won his first Kaggle competition

tom wins kaggle

This is a copy of Tom’s original post on Github.

Winning approach of the Facebook V Kaggle competition

The Facebook V: Predicting Check Ins data science competition where the goal was to predict which place a person would like to check in to has just ended. I participated with the goal of learning as much as possible and maybe aim for a top 10% since this was my first serious Kaggle competition attempt. I managed to exceed all expectations and finish 1st out of 1212 participants! In this post, I’ll explain my approach.

Overview

This blog post will cover all sections to go from the raw data to the winning submission. Here’s an overview of the different sections. If you want to skip ahead, just click the section title to go there.

The R source code is available on GitHub. This thread on the Kaggle forum discusses the solution on a higher level and is a good place to start if you participated in the challenge.

Introduction

Competition banner

Competition banner

From the competition page: The goal of this competition is to predict which place a person would like to check in to. For the purposes of this competition, Facebook created an artificial world consisting of more than 100,000 places located in a 10 km by 10 km square. For a given set of coordinates, your task is to return a ranked list of the most likely places. Data was fabricated to resemble location signals coming from mobile devices, giving you a flavor of what it takes to work with real data complicated by inaccurate and noisy values. Inconsistent and erroneous location data can disrupt experience for services like Facebook Check In.

The training data consists of approximately 29 million observations where the location (x, y), accuracy, and timestamp is given along with the target variable, the check in location. The test data contains 8.6 million observations where the check in location should be predicted based on the location, accuracy and timestamp. The train and test data set are split based on time. There is no concept of a person in this dataset. All the observations are events, not people.

A ranked list of the top three most likely places is expected for all test records. The leaderboard score is calculated using the MAP@3 criterion. Consequently, ranking the actual place as the most likely candidate gets a score of 1, ranking the actual place as the second most likely gets a score of 1/2 and a third rank of the actual place results in a score of 1/3. If the actual place is not in the top three of ranked places, a score of 0 is awarded for that record. The total score is the mean of the observation scores.

Check Ins where each place has a different color

Check Ins where each place has a different color

Exploratory analysis

Location analysis of the train check ins revealed interesting patterns between the variation in x and y. There appears to be way more variation in x than in y. It was suggested that this could be related to the streets of the simulated world. The difference in variation between x and y is however different for all places and there is no obvious spatial (x-y) pattern in this relationship.

It was quickly established by the community that time is measured in minutes and could thus be converted to relative hours and days of the week. This means that the train data covers 546 days and the test data spans 153 days. All places seem to live in independent time zones with clear hourly and daily patterns. No spatial pattern was found with respect to the time patterns. There are however two clear dips in the number of check ins during the train period.

Accuracy was by far the hardest input to interpret. It was expected that it would be clearly correlated with the variation in x and y but the pattern is not as obvious. Halfway through the competition I cracked the code and the details will be discussed in the Feature engineering section.

I wrote an interactive Shiny application to research these interactions for a subset of the places. Feel free to explore the data yourself!

Problem definition

The main difficulty of this problem is the extended number of classes (places). With 8.6 million test records there are about a trillion (10^12) place-observation combinations. Luckily, most of the classes have a very low conditional probability given the data (x, y, time and accuracy). The major strategy on the forum to reduce the complexity consisted of calculating a classifier for many x-y rectangular grids. It makes much sense to make use of the spatial information since this shows the most obvious and strong pattern for the different places. This approach makes the complexity manageable but is likely to lose a significant amount of information since the data is so variable. I decided to model the problem with a single binary classification model in order to avoid to end up with many high variance models. The lack of any major spatial patterns in the exploratory analysis supports this approach.

Strategy

Generating a single classifier for all place-observation combinations would be infeasible even with a powerful cluster. My approach consists of a stepwise strategy in which the conditional place probability is only modeled for a set of place candidates. A simplification of the overall strategy is shown below

High level strategy

High level strategy

The given raw train data is split in two chronological parts, with a similar ratio as the ratio between the train and test data. The summary period contains all given train observations of the first 408 days (minutes 0-587158). The second part of the given train data contains the next 138 days and will be referred to as the train/validation data from now on. The test data spans 153 days as mentioned before.

The summary period is used to generate train and validation features and the given train data is used to generate the same features for the test data.

The three raw data groups (train, validation and test) are first sampled down into batches that are as large as possible but can still be modeled with the available memory. I ended up using batches of approximately 30,000 observations on a 48GB workstation. The sampling process is fully random and results in train/validation batches that span the entire 138 days’ train range.

Next, a set of models is built to reduce the number of candidates to 20 using 15 XGBoost models in the second candidate selection step. The conditional probability P(place_match|features) is modeled for all ~30,000*100 place-observation combinations and the mean predicted probability of the 15 models is used to select the top 20 candidates for each observation. These models use features that combine place and observation measures of the summary period.

The same features are used to generate the first level learners. Each of the 100 first level learners are again XGBoost models that are built using ~30,000*20 feature-place_match pairs. The predicted probabilities P(place_match|features) are used as features of the second level learners along with 21 manually selected features. The candidates are ordered using the mean predicted probabilities of the 30 second level XGBoost learners.

All models are built using different train batches. Local validation is used to tune the model hyperparameters.

Candidate selection 1

The first candidate selection step reduces the number of potential classes from >100K to 100 by considering nearest neighbors of the observations. I considered the neighbor counts of the 2500 nearest neighbors where y variations are 2.5 times more important than x variations. Ties in the neighbor counts are resolved by the mean time difference since the observations. Resolving ties with the mean time difference is motivated by the shifts in popularity of the places.

The nearest neighbor counts are calculated efficiently by splitting up the data in overlapping rectangular grids. Grids are created as small as possible while still guaranteeing that the 2500 nearest neighbors fall within the grid in the worst case scenario. The R code is suboptimal through the use of several loops but the major bottleneck (ordering the distances) was reduced by a custom Rcpp package which resulted in an approximate 50% speed up. Improving the logic further was no major priority since the features were calculated on the background.

Feature engineering

Feature engineering strategy

Three weeks into the eight-week competition, I climbed to the top of the public leaderboard with about 50 features. Ever since I kept thinking of new features to capture the underlying patterns of the data. I also added features that are similar to the most important features in order to capture the subtler patterns. The final model contains 430 numeric features and this section is intended to discuss the majority of them.

There are two types of features. The first category relates to features that are calculated using only the summary data such as the number of historical check ins. The second and largest category combines summary data of the place candidates with the observation data. One such example is the historical density of a place candidate, one year prior to the observation.

All features are rescaled if needed in order to result in similar interpretations for the train and test features.

Location

The major share of my 430 features is based on nearest neighbor related features. The neighbor counts for different Ks (1, 5, 10, 20, 50, 100, 250, 500, 1000 and 2500) and different x-y ratio constants (1, 2.5, 4, 5.5, 7, 12 and 30) resulted in 10*7 features. For example: if a test observation has 3 of its 5 nearest neighbors of class A and 2 of its 5 nearest neighbors as class B, the candidate A will contain the numeric value of 3 for the K=5 feature, the candidate B will contain the numeric value of 2 for the K=5 feature and all other 18 candidates will contain the value of 0 for that feature. The mean time difference between a candidate and all 70 combinations resulted in 70 additional features. 10 more features were added by considering the distance between the Kth features and the observations for a ratio constant of 2.5. These features are an indication of the spatial density. 40 more features were added in a later iteration around the most significant nearest neighbor features. K was set at (35, 75, 100, 175, 375) for x-y ratio constants (0.4, 0.8, 1.3, 2, 3.2, 4.5, 6 and 8). The distances of all 40 combinations to the most distant neighbor were also added as features. Distance features are divided by the number of summary observations in order to have similar interpretations for the train and test features.

I further added several features that consider the (smoothed) spatial grid densities. Other location features relate to the place summaries such as the median absolute deviations and standard deviations in x and y. The ratio between the median absolute deviations was added as well. Features were relaxed using additive (Laplace) smoothing with different relaxation constants whenever it made sense using the relaxation constants 20 and 300. Consequently, the relaxed mad for a place with 300 summary observation is equal to the mean of the place mad and the weighted place population mad for a relaxation constant of 300.

Time

The second largest share of the features set belongs to time features. Here I converted all time period counts to period density counts in order to handle the two drops in the time frequency. Periods include 27 two-week periods prior to the end of the summary data and 27 1-week periods prior to the end of the summary data. I also included features that look at the two-week densities looking back between 75 and 1 weeks from the observations. These features resulted in missing values but XGBoost is able to handle them. Additional features were added for the clear yearly pattern of some places.

Weekly counts

Weekly counts

Hour, day and week features were calculated using the historical densities with and without cyclical smoothing and with or without relaxation. I suspected an interaction between the hour of the day and the day of the week and also added cyclical hour-day features. Features were added for daily 15-minute intervals as well. The cyclical smoothing is applied with Gaussian windows. The windows were chosen such that the smoothed hour, hour-week and 15-minute blocks capture different frequencies.

Other time features include extrapolated weekly densities using various time series models (arima, Holt-Winters and exponential smoothing). Further, the time since the end of the summary period was also added as well as the time between the end of the summary period and the last check in.

Accuracy

Understanding accuracy was the result of generating many plots. There is a significant but low correlation between accuracy and the variation in x and y but it is not until accuracy is binned in approximately equal sizes that the signal becomes visible. The signal is more accurate for accuracies in the 45-84 range (GPS data?).

Mean variation from the median in x versus 6 time and 32 accuracy groups

Mean variation from the median in x versus 6 time and 32 accuracy groups

The accuracy distribution seems to be a mixed distribution with three peaks which changes over time. It is likely to be related to three different mobile connection types (GPS, Wi-Fi or cellular). The places show different accuracy patterns and features were added to indicate the relative accuracy group densities. The middle accuracy group was set to the 45-84 range. I added relative place densities for 3 and 32 approximately equally sized accuracy bins. It was also discovered that the location is related to the three accuracy groups for many places. This pattern was captured by the addition of additional features for the different accuracy groups. A natural extension to the nearest neighbor calculation would incorporate the accuracy group but I did no longer have time to implement it.

The x-coordinates seem to be related to the accuracy group for places like 8170103882

The x-coordinates seem to be related to the accuracy group for places like 8170103882

Z-scores

Tens of z scores were added to indicate how similar a new observation is to the historical patterns in the place candidates. Robust Z-scores ((f-median(f))/mad(f) instead of (f-mean(f))/sd(f)) gave the best results.

Most important features

Nearest neighbors are the most important features for the studied models. The most significant nearest neighbor features appear around K=100 for distance constant ratios around 2.5. Hourly and daily densities were all found to be very important as well and the highest feature ranks are obtained after smoothing. Relative densities of the three accuracy groups also appear near the top of the most important features. An interesting feature that also appears at the top of the list relates to the daily density 52 weeks prior to the check in. There is a clear yearly pattern which is most obvious for places with the highest daily counts.

Clear yearly pattern for place 5872322184. The green line goes back 52 weeks since the highest daily count

Clear yearly pattern for place 5872322184. The green line goes back 52 weeks since the highest daily count

The feature files are about 800MB for each batch and I saved all the features to an external HD.

Candidate selection 2

The features from the previous section are used to generate binary classification models on 15 different train batches using XGBoost models. With 100 candidates for each observation, this is a slow process and it made sense to me to narrow down the number of candidates to 20 at this stage. I did not perform any downsampling in my final approach since all zeros (not a match between the candidate and true match) contain valuable information. XGBoost is able to handle unbalanced data quite well in my experience. I did however consider to omit observations that didn’t contain the true class in the top 100 but this resulted in slightly worse validation scores. The reasoning is the same as above: those values contain valuable information! The 15 candidate selection models are built with the top 142 features. The feature importance order is obtained by considering the XGBoost feature importance ranks of 20 models trained on different batches. Hyperparameters were selected using the local validation batches. The 15 second candidate selection models all generate a predicted probability of P(place_match|data), I average those to select the top 20 candidates in the second candidate selection step.

At this point I also dropped observations that belong to places that only have observations in the train/validation period. This filtering was also applied to the test set.

First level learners

The first level learners are very similar to the second candidate selection models other than the fact that they were fit on one fifth of the data for 75 of the 100 models. The other 25 models were fit on 100 candidates for each observation. The 100 base XGBoost learners were fit on different random parts of the training period. Deep trees gave me the best results here (depth 11) and the eta constant was set to (11 or 12)/500 for 500 rounds. Column sampling also helped (0.6) and subsampling the observations (0.5) did not hurt but of course resulted in a fitting speed increase. I included either all 430 features or a uniform random pick of the ordered features by importance in a desirable feature count range (100-285 and 180-240). The first level learner framework was created to handle multiple first level learner types other than XGBoost. I experimented with the nnet and H2O neural network implementations but those were either too slow in transferring the data (H2O) or too biased (nnet). The way XGBoost handles missing values is another great advantage over the mentioned neural network implementations. Also, the XGBoost models are quite variable since they are trained on different random train batches with differing hyperparameters (eta constant, number of features and the number of considered candidates (either 20 or 100)).

Second level learners

The 30 second level learners combine the predictions of the 100 first level models along with 21 manually selected features for all top 20 candidates. The 21 additional features are high level features such as the x, y and accuracy values as well as the time since the end of the summary period. The added value of the 21 features is very low but constant on the validation set and the public leaderboard (~0.02%). The best local validation score was obtained by considering moderate tree depths (depth 7) and the eta constant was set to 8/200 for 200 rounds. Column sampling also helped (0.6) and subsampling the observations (0.5) did not hurt but again resulted in a fitting speed increase. The candidates are ordered using the mean predicted probabilities of the 30 second level XGBoost learners.

Analysis of the local MAP@3 indicated better results for accuracies in the 45-84 range. The difference between local and test validation scores is in large part related to this observation. There seems to be a trend towards the use of devices that show less variation .

Local MAP@3 versus accuracy groups

Local MAP@3 versus accuracy groups

Conclusion

The private leaderboard standing below, used to rank the teams, shows the top 30 teams. It was a very close competition in the end and Markus would have been a well-deserved winner as well. We were very close to each other ever since the third week of the eight-week contest and pushed each other forward. The fact that the test data contains 8.6 million records and that it was split randomly for the private and public leaderboard resulted in a very confident estimate of the private standing given the public leaderboard. I was most impressed by the approaches of Markus and Jack (Japan) who finished in third position. You can read more about their approaches on the forum. Many others also contributed valuable insights.

Private leaderboard score (MAP@3) - two teams stand out from the pack

Private leaderboard score (MAP@3) – two teams stand out from the pack

I started the competition using a modest 8GB laptop but decided to purchase a €1500 workstation two weeks into the competition to speed up the modeling. Starting with limited resources ended up to be an advantage since it forced me to think of ways to optimize the feature generation logic. My big friend in this competition was the data.table package.

Running all steps on my 48GB workstation would take about a month. That seems like a ridiculously long time but it is explained by the extended computation time of the nearest neighbor features. While calculating the NN features I was continuously working on other parts of the workflow so speeding the NN logic up would not have resulted in a better final score.

Generating a ~.62 score could however be achieved in about two weeks by focusing on the most relevant NN features. I would suggest to consider 3 of the 7 distance constants (1, 2.5 and 4) and omit the mid KNN features. Cutting the first level models from 100 to 10 and the second level models from 30 to 5 would also not result in a strong performance decrease (estimated decrease of 0.1%) and cut the computation time to less than a week. You could of course run the logic on multiple instances and further speed things up.

I really enjoyed working on this competition even though it was already one of the busiest periods of my life. The competition was launched while I was in the middle of writing my Master’s Thesis in statistics in combination with a full time job. The data shows many interesting noisy and time dependent patterns which motivated me to play with the data before and after work. It was definitely worth every second of my time! I was inspired by the work of other Kaggle winners and successfully implemented my first two level model. Winning the competition is a nice extra but it’s even better to have learnt a lot from the other competitors, thank you all!

I look forward to your comments and suggestions, please go to my original post to post your comments.

Tom.

How to innovate in the Age of Big Data presented by Stephen Brobst

stephen3

Executive Summer Session

Stephen Brobst will be at the European Data Innovation Hub. We asked him to share his views on the importance of open data, open source, analytics in the cloud and data science. Stephen is on the forefront of the technology and we can’t wait to hear what is happening in the Silicon Valley. Count on it that you will leave the workshop inspired and weaponed with some actionable ideas that will help us to define a profitable strategy for the data science teams.

Format of the session :

  • 15:00 – Keynote:How to innovate in the Age of Big Data
  • 15:50 – Open Discussion on “Sustainable Strategies for Data Science, tackling following topics:
  • Data Science is the Key to Business Success
  • Three Critical Technologies Necessary for Big Data Exploitation
  • How to Innovate in the Age of Big Data
  • 16:45 – Networking Session

Stephen Brobst is the Chief Technology Officer for Teradata Corporation.  Stephen performed his graduate work in Computer Science at the Massachusetts Institute of Technology where his Masters and PhD research focused on high-performance parallel processing. He also completed an MBA with joint course and thesis work at the Harvard Business School and the MIT Sloan School of Management.  Stephen is a TDWI Fellow and has been on the faculty of The Data Warehousing Institute since 1996.  During Barack Obama’s first term he was also appointed to the Presidential Council of Advisors on Science and Technology (PCAST) in the working group on Networking and Information Technology Research and Development (NITRD).  He was recently ranked by ExecRank as the #4 CTO in the United States (behind the CTOs from Amazon.com, Tesla Motors, and Intel) out of a pool of 10,000+ CTOs.

Job – Junior Data Scientist

Screenshot 2016-07-01 12.02.02

Are you pursuing a career in data science?

We have a great opportunity for you: an intensive training program combined with interesting job opportunities!

Interested? Check out http://di-academy.com/bootcamp/ follow the link to our datascience survey and send your cv to training@di-academy.com

Once selected, you’ll be invited for the intake event that will take place in Brussels this summer.

Hope to see you there,

Nele & Philippe