Figure 1.2: Basic setup of the chooses g that best matches fon the tra with the

Question

Figure 1.2: Basic setup of the chooses g that best matches fon the tra with the hope that it will continue to or not this hope is justihed remains to be training example of evi match f on new vious custou omers. W 2 illustrates et Hthro unctional components of the learning problem. The weight ult is co credit. s Exercise 1.I Express each of the following tasks in the specifying the input space T, output space framework of lea: the specifics of tine A patient walks in with a nedical history and som ce ', target fu the data set that we will learn from and the specifics of from d s data by u want to identify the problem recognition (for example postal zip code recognit symptoms, and yo for mail sorting) This formula c (c) Determining if an email is spam how an electric load varies with price, temperature, and bliem of interest to you for which there is no analytic solution or not day of the week Aroeda rm which to construct an empirical soluticn where ai, prove credi sign(s) determine but you have data f This of artifi The

Explanation / Answer

Medical diagnosis :

X : Medical history and some symptoms
Y : The problem we want to identify
Target function : Diagnosing the problem ie whether the problem has occured or not
Data type : Some numerical values of tests of medical history ( to be used as perdictors ) and some qualitative or quantitave data on symptoms occuring to the patient.

Digit Recognition :

X : Images of digits
Y : Any of the 10 digits from 0 to 9
Target function : Detecting the digits from images
Data type : Scanned images of digits to be identified and some labeled images of digits to train the model.

Email spam or not :

X : Emails
Y : Binary output - spam or not
Target function : Naive Bayes classifier
Data type : Content of emails to be identified as spam or not and some labeled emails to train the model.

Electric load prediction :

X : price, temperature and day of week
Y : Electric load
Target function : Regression
Data type : Data on electric load and its corresponding price, temperature and day of work. It will be used to fit the regression model.

Random problem :

X : Unstructued data
Y : Unstructued data
Target function : Unsupervised learning
Data type : Unstructued data

Related Questions

Navigate

• Browse (All)
• Browse F
• Subjects

---

---

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.