Essay Example on Spam Detection Spam via emails SMS or any other medium is very dangerous

Subcategory:

Category:

Words:

448

Pages:

2

Views:

202
Spam Detection Spam via emails SMS or any other medium is very dangerous as it may cause loss of money or loss of other important belongings or leaking of important data Spam is very dangerous to any individual or organization due to its nature of loss of capita and information leaking Some spams may even inject malware into the system only by just opening them Spam detection works by previously reported or flag ed or by the system itself by analysing the contents How will you choose the right algorithm Machine learning algorithm is categorized in three types they are Classification Regression and Clustering An email may be either spam or will be not so classification is the best choice for this as it should fit in either class spam or not spam Write in detail about the different steps that you will perform to develop the machine learning algorithm Collect Data Data is collected from previous mails from email clients that are already sorted Analyze Data This data is analyzed to understand the information available like what makes an email to be spam etc Prepare Data Having a proper dataset with proper emails as some incomplete emails without any meaning must be removed or information must be added so as to improve the dataset for the purpose it is to be used for

Only data which will help to detect whether it is spam or normal email must be used i e useless data must be removed and take data which would improve results Train and Test This data will be trained using 80 of training set and tested using 20 of the training set Automated Braking Automated Braking may be part of a whole system Automated driving or an independent part where only braking is automated Brake Assist How will you choose the right algorithm Braking is not like either full braking 100 and coming to stop or no braking 0 at all Braking have multiple values which may be applied according to the present scenario Hence regression is the best machine learning algorithm for automated braking Here the braking system will use input sensors for detecting signals where full brake will be applied to bring vehicle to standstill while for other cases where an obstacle comes say like speed braker road bumps where the vehicle should be brought under a certain speed only etc Write in detail about the different steps that you will perform to develop the machine learning algorithm Collect Data Data is collected from manufacturer about braking etc for accuracy Analyze Data Study parameters and their relations that affect braking of a vehicle like speed type of brake braking power type of surface road type of tyre friction value etc Prepare Data Dataset in the form of regression tree or CART where various factor mentioned above with value ranges and the target output as how much braking would be needed 



As braking power is not calculated in time since braking will eventually led to stopping of the vehicle which would be not useful on road bumps so instead output would be required in terms of speed like for speed bumps 25kmph for red signals and other emergency braking etc 0 kmph etc Train and Test This data will be trained using 80 of training set and tested using 20 of the training set Stock Market Prediction Stock market prediction is very difficult job as wrong prediction may lead to loss of capital As stock market depends say of a particular company organization on its performance current and previous decisions taken and their effects like cancelling a product can also lead to low stock price etc as this factor influence investors to either invest or withdraw shares How will you choose the right algorithm As stock market is dependent on various factors and predicting if price per share would rise or fall and by how much would require regression due to numeric prediction Write in detail about the different steps that you will perform to develop the machine learning algorithm Collect Data By mapping the stock price by the date and time Analyze Data Analyzing the factors that cause people to invest or withdraw that results in shares to rise or fall Prepare Data Dataset will be then cleaned by removing incomplete data or adding missing information where there is missing price on a particular date and time etc Train and Test

This data will be trained using 80 of training set and tested using 20 of the training set Recommendations Recommendations anywhere online depends on previously searched or visited content They show same or similar content with prices and discounts or other benefits How will you choose the right algorithm Recommendations track our previous sessions and suggest accordingly it may recommend similar product or product which is likely to be purchased Recommendation may be from same category say book from sci fi genre so recommendation may give other sci fi books as well as books from another genre too So suggesting from other genre would mean clustering as it tries to match checks for similarity for fitting data Write in detail about the different steps that you will perform to develop the machine learning algorithm Collect Data By similar products together history i e what user check after checking a product Analyze Data Learning how people would most likely check the recommendation from the data provided Prepare Data Cleaning the data by selecting data which is more likely to be seen by the user after it is recommended i e the recommendation is not ignored by the user Train and Test This data will be trained using 80 of training set and tested using 20 of the training set

Write and Proofread Your Essay
With Noplag Writing Assistance App

Plagiarism Checker

Spell Checker

Virtual Writing Assistant

Grammar Checker

Citation Assistance

Smart Online Editor

Start Writing Now

Start Writing like a PRO

Start