1.2 Support Vector Machine (SVM)SVM is a supervised methodology rigorously based on the statistical learning theory14.For linearly separable examples, SVM constructs a maximum margin hyperplaneseparating the data points into two di erent classes. This hyperplane acts as a decisionsurface between the two classes. For nonlinearly separable data, SVM rsttransforms the data into a higher dimensional feature space and subsequently employsa linear maximum margin hyperplane shown in Figure.1.3.This may introducea computational intractability requiring a transformation to high dimensional space.SVM handles this by de ning appropriate kernel functions by virtue of which thecomputations can be carried out in the original space itself. Three popular kernelfunctions are Linear, Polynomial and Radial Basis Function (RBF). In bioinformat-ics, many domain speci c kernel functions are also available like graph kernel andstring kernel. The concept can be extended to multiclass classi cation. Two popularmulticlass classi cation methods are employed viz., one against all and one againstone. The general steps involved in the SVM algorithm are as follows: Construction of a feature vector representing the positive and negative dataset:This feature vector consists of the properties of the input data like amino acidand/or dipeptide composition, physico-chemical properties etc. Choice of an appropriate kernel function suitable for the prediction task usingthe classi er training Training of SVM classi er by selecting optimum kernel parameters so as toachieve highest accuracy Selection of model with the best performance to perform predictions Application of selected model for performing predictions on the unknown inputdata setFigure 1.3: OPTIMAL SEPARATION HYPER-PLAINSVM is the most robust classi er, and has the best generalization ability on theunseen data as compared to other methods. It is the most commonly used machinelearning method in bioinformatics andcomputational biology. It has been employedfor secondary structure prediction, fold recognition, bindingsite prediction as wellas for gene nding 15136.5

·     
Safety Violations:

 

BP had a record of non-compliance with
environmental regulations and safety of workers much before the Deepwater
Horizon accident. In 2002, BP was charged with $100 million for falsification
of inspection report of a refinery in Los Angeles. In 2005, an explosion
occurred at a Texas refinery causing deaths and casualties. BP was fined with
$87 million for noncompliance with safety regulations. In 2006, it was
discovered that BP’s Alaska pipeline was leaking, despite replacement, the leak
still continued. Reports and investigations suggest that various shortcuts were
taken with respect to the drilling of the Macondo oil field where the accident
took place. An internal report of BP provides details of ten near miss
accidents in the Gulf of Mexico in 2007.

We Will Write a Custom Essay Specifically
For You For Only $13.90/page!


order now

 

Three red flags and warnings had been
issued to BP. The main reason for the accident was the usage of the dangerous
single well casing, fewer devices installed to center the drill pipe,
recommended centralizers were not purchased and mud was not circulated through
the well to test the quality of the cement barrier. They even failed the
pressure test that was conducted on the site. BP violated industry standards
and took decisions even being warned by the employees and subcontractors. Rigs
continued to function under risk. BP did not take any action towards the
maintenance of the rigs and ignored red flags and warnings from the procedural
test itself. In short, they compromised the safety of workers in the name of
cost savings.