Posted milena-piryankova, stephen, nike, bogomil-filipov, vasilmarchev, martinmarinov, sergeyvi4ev, michael-opitsch, penchodobrev, sist Mentors: junior, pepe1 CommentPosted in Prediction systemsPrediction systems

Datathon Kaufland Solution – Team Total Kaputt! – Why da faQ the machine broke down?

What we tried to do to solve the Kaufland case for the Global Datathon 2018. This article just contains our exploratory data analysis in the form of many plots and some explanations. There isn’t any modeling stage described here.

Posted junior, zenpanik, alex-efremov, pepe, tomislavk, liad, preslav, tony, laura, mladensavov, anenov, metodinikolov, drceenish, leomrsic, boryana, mcneto, jjagusch, drsubhabahapal, sist, jans, gkmitov, metodinikolov, gloria, quintshekharLeave a commentPosted in GD2018 Mentors, MentorsGD2018 Mentors, Mentors

Datathon Ontotext Mentors’ Guidelines – Text Mining Classification

In this article the mentors give some preliminary guidelines, advice and suggestions to the participants for the case. Every mentor should write […]

Posted sujit, sujit, prakash, svarya, vijays, amit98 Mentors: drsubhabahapal6 CommentsPosted in Datathons Solutions, Learn, Team solutionsDatathons Solutions, Learn, Team solutions

Cryptocurrency Prediction by Kautilya

Given the cryptocurrencies’ data, we aim to forecast the future cryptocurrencies’ prices so as to execute profitable trades. We show that the cryptocurrencies’ prices also exhibit desirable properties such as stationarity and mixing. Some classical time series prediction models that exploit this behavior, such as “Arima” models produce poor predictions and also lack good probabilistic interpretations. We have introduced a theoretical framework in the 1st place and for predicting and trading prices of the cryptocurrencies for future and based on that we have designed our model which is based on “Neural Network” model which can give better prediction values as compared to the other models.

Posted vsenderov, valan, hvrigazov, cecopld, mitzev, tonypetrov, pepe Mentors: tony6 CommentsPosted in Image recognition, Learn, Team solutionsImage recognition, Learn, Team solutions

Team Cherry. The Kaufland Case. Fast and Accurate Image Classification Architecture for Recognizing Produce in a Real-Life Groceries’ Setting

Our best model (derived from VGG) achieved 99.46% top3 accuracy (90.18% top1) with processing time during training of 0.006 s per image on a single GPU Titan X (200s / epoch with 37 000 images).

The teams vision is for the team members to see where they stand compared to others in terms of ideas and approaches to computer vision and to learn new ideas and approaches from the other team-mates and the mentors.

Therefore the team is pursuing a pure computer vision approach to solving the Kaufland and/or the ReceiptBank cases.