1. Assignment Of Data Science
No#4
Submitted To:
Sir Ehtisham Rasheed
Submitted By:
Ihtesham Tariq (16160819-032)
M.Amir Zeshaan (16160819-016)
Mubeen Asghar (16160819-002)
MSc Computer Science
4th Semester
UNIVERSITY OF GUJTRAT G.T ROAD
CAMPUS
2. Q1. Give an example of machine learning that you see in everyday life. In
what ways has it been helpful or how do you think it can be improved?
Answer:
Frompersonalizing your news feed to better ads targeting, social media platforms
are utilizing machine learning for their own and user benefits. Here are a few
examples that you must be noticing, using, and loving in your social media
accounts, withoutrealizing that these wonderfulfeatures are nothing but the
applications of ML.
People You May Know: Machine learning works on a simple concept:
understanding with experiences. Facebook continuously notices the friends that
you connect with, the profiles that you visitvery often, your interests, workplace,
or a group that you sharewith someoneetc. On the basis of continuous learning,
a list of Facebook users are suggested that you can become friends with.
Face Recognition: You upload a picture of you with a friend and Facebook
instantly recognizes that friend. Facebook checks the poses and projections in the
picture, notice the unique features, and then match them with the people in your
friend list. The entire process atthe backend is complicated and takes care of the
precision factor but seems to be a simple application of ML at the front end.
Similar Pins: Machine learning is the core element of Computer Vision, which is a
technique to extract useful information fromimages and videos. Pinterest uses
computer vision to identify the objects (or pins) in the images and recommend
similar pins accordingly.
Smartmachines and applications are steadily becoming a daily phenomenon,
helping us make faster, moreaccurate decisions.
And with morethan 75 percent of businesses investing in Big Data, the role of AI
and machine learning is set to increase dramatically over the next five years.
As of 2017, a quarter of organisations arespending 15 percent or more of their IT
budget on machine learning capabilities, and we expect the number of machine
learning examples to rise in the near future.
3. Q2. In building a machine learning model, why do we want to adjust the
parameters?
a) To reduce the model’s error
b) To compare different model variations
c) To provide the best graphof the model outputs
Answer:
In building a machine learning model wewant to adjustthe parameters like “to
reduce the model’s error” , ”To compare different model variations” and “to
providethe best graph of model outputs” because its essential for build model for
machine learning. For following rules we become build accurate models without
errors , variation, and wrong output.
Q3:What is the first stepinconstructing adecisiontree?
a) Start withall samples at a node
b) Repeatedly partitiondataintosuccessively purer subsetsuntil
stopping criteriaare satisfied.
c) Partitionthe samples intosubsets basedonthe input variables.
Answer:
Repeatedly partition data into successively proper subsets until stopping criteria are satisfied.
Q4. What is the command to get the number of rows in a data set titled“data”?
a) data.shape[0]
b) data.shape[1]
c) data.size()
d) data.length()
Answer:
“Data.shape” is useto get no of rows in data set titled “data”.
4. Q5. Describe anatural language processing applicationthat youhave seenor
used. Explain how data can improve the accuracy and time efficiency of this
everyday application.
Answer:
In natural languageprocessing application we haveseen that we can easily
cleaning data or text and reduce the size counting in limited time and easily
performanalysis in shorttime.
Using python.
Itis help full in our life when we havelarge text and we want to find statistics
then we use natural language processing.
Q6. There were news articles in the recent past of a major company filtering
content for users based on sentiment to evaluate the impact on the users. How
might this be used in a good, or bad, way?
Answer:
Twitter is recently filtering content for users based on sentiments to
evaluate the impact on the users. It is “good way” because its helpful for further
studies on human sentiments.
Q7. Given that you now have accessto twitter data, is there a question you
would be particularly interestedin asking ofthe data? How would you find
the information and what canyou analyze from it?
Answer:
If we have access the twitter data we will have a question that most effected
trends from “Social”, “Political” and “religious”. We can easily find calution and
analysis from data after getting some specific data of trends and their rating
accordingly their names or genrs. We can analyze that what is trend type is mostly
reading and popular on twitter.