Search algorithms are designed to explore spaces systematically.
Real world problem are more complex.
We can use iterative improvement algorithms and local search.
The objective is to find best state according to an optimization function.
Examples such as:
Integrated circuit design, telecommunications network optimization.
Idea: Keep a single “current state”, and try to improve it.
Move only to neighbors of that node
1) No need to maintain a search tree
2) Use very little memory
3) Can often find good enough solutions in continuos or large state spaces.
Local Search Algorithms:
1) Hill climbing (steepest ascent / descent)
2) simulated annealing: inspired by statistical physics
3) Local beam search
4) Genetic algorithm: inspired by evolutionry biology
Data come in different sizes and also flavors (types):
9) Some or all of the above
The fact is that today we are datafied.
Wherever we go, we leave a trail of data.
Data science aims to make a good use of the data for good sake of humanity
The data science process consists of five main steps:
1) Data colection
2) Data preparation (to be feed into machine learning)
3) Exploratory data analysis (is it enough? Do we need to collect more)
4) Machine learning (model) -> prediction in making better data driven decison
Machine learning can be apply in many area with example as below:
-digit recognition on checks, zip codes
-detection faces in images
-MRI image analysis
-credit card fraud detection
With some examples above, is there any potential companies turn up in your mind that involves themselves in relevant business?