Page 53 - Contributed Paper Session (CPS) - Volume 6

CPS1490 Nehall Ahmed Farouk Mohamed

Considering the work of the team in the project and the methods that was
used to overcome ML problems in big data prediction models, the model's
basic framework illustrated in figure (3).

3. Results
Solving the challenges of ML can be progressed through dealing with each
problem from its perspective. Missing data and incompleteness can be solved
before start working on the predictive mode, by labelling the data and
estimating the missing values. Also the main problem of the huge and massive
size of big data can be solved through using fine grain technique, where it is
divided or classified. Parallel algorithms for data and models helped a lot in
improving the running time complexity and model complexity. Training a

42 | I S I W S C 2 0 1 9