代做COMPSCI5100 Machine Learning & Artificial Intelligence for Data Scientists 22-23调试Haskell程序

DEGREES OF MSc, MSci, MEng, BEng, BSc,MA and MA (Social Sciences)

COMPSCI5100

Machine Learning & Artificial Intelligence for Data Scientists

Question 1: Regression (Total marks: 20)

Consider using regression to predict the world population growth rate using the data shown in the following figure:

Figure 1. World population growth rate from 1950 to 2021. Source: https://ourworldindata.org/worldpopulation–update–2022

(a) Propose a rescaling strategy (with enough details of the procedure) for the variable Year. Explain why the proposed strategy is appropriate.                [4 marks]

(b) Consider fitting the data with a polynomial regression model with the order of 1,

identify the two most likely poorly fitted data points and explain why.           [6 marks]

(c) Consider fitting the data in figure 1 with a regression with the sigmoid basis function:

where xn  represents each year and sigmoid(a) = 1/(1+exp(-a)). Outline one advantage and disadvantage of using this sigmoid basis function over polynomials. [4 marks]

(d) Suppose we use the sigmoid basis function in (c), with μk  set to be xn  and  s  = 10, to to fit the data. We used three fitting strategies, namely linear regression, ridge regression and lasso, and obtained the following fitting model in Figure 2 A, B and C. Identify which fitting strategy is used in each figure and explain why (note, each method is used only once). [6 marks]

Figure 2 A

Figure 2 B

Figure 2 C

Question 2: Classification (Total marks: 20)

(a) You have been asked to design a classifier to automatically identify the Tweets that are considered as ‘hate speech’ in the social media website Twitter. You collected a training dataset which has 800 ‘regular’ tweets and 100 ‘hate’ tweets. Answer the following:

(i) Describe 2 features you might use for this task including their type (scalar/vector,  real- valued or not). [3 marks]

(ii) You learn a faulty classifier which always classifies a tweet as ‘regular’. What would be the weighted classification accuracy of this classifier? [2 marks]

(iii) Assume that we use Logistic Regression for the task considering only 2 scalar features. We get the following set of parameters w = [-1.8, 2.1 -0.3]T  after training. For feature vector x = [1, 1]T, calculate the output of the logistic function (probability score). [4 marks]

(iv)  Logistic Regression assumes a linear relationship between the dependent and the independent variables (x). Why is that considered a limitation of the model? [2 marks]

(v) Can we replace the sigmoid function in Logistic Regression by the function g(z) shown in Fig. 3? Explain your answer. [2 marks]

(b) Consider the 8 data points for your training set belonging to 2 classes in Fig. 4. This is used to train a linear SVM.

(i) Draw the decision boundary for linear hard margin SVM method with a solid line. Show the margin using dotted lines. [3 marks]

(ii) Which ones are the support vectors? [2 marks]

(iii) What is the training error? [1 mark]

(iv) Removal of which data point will change the decision boundary? [1 mark]

3. Unsupervised learning question (Total marks 20)

Consider using the K-means algorithm to perform. clustering on the following scenario A1.

We expect to form. two clusters as shown in A2.

(a) Outline what would happen if we directly apply K-means with Euclidean distance to this data. Can it achieve the clustering objective? How will it split/group the data and why? [2 marks]

(b) An alternative approach is to use Kernel K-means. Would kernel K-means could help in this dataset and why? [3 marks]

(c) An alternative approach is to use mixture models. Would mixture models help to better classify this dataset than K-means and why? [3 marks]

(d) The plot in D1 shows some 2D data. PCA is applied to this data. Sketch this plot, and indicate on your sketch what the first principal component would look like. Explain your reasoning.

D1) 2D points [2 marks]

(e) Similar to previous question, sketch what the second principal component would look like and explain why.        [2 marks]

(f) Explain why PCA is used (provide at least two applications) and how to decide for the optimum number of principal components.           [2 marks]

(g) Explain the advantages and disadvantages of feature selection based on projection compared to feature selection based on how well they can discriminate between two classes.                [6 marks]


热门主题

课程名

mktg2509 csci 2600 38170 lng302 csse3010 phas3226 77938 arch1162 engn4536/engn6536 acx5903 comp151101 phl245 cse12 comp9312 stat3016/6016 phas0038 comp2140 6qqmb312 xjco3011 rest0005 ematm0051 5qqmn219 lubs5062m eee8155 cege0100 eap033 artd1109 mat246 etc3430 ecmm462 mis102 inft6800 ddes9903 comp6521 comp9517 comp3331/9331 comp4337 comp6008 comp9414 bu.231.790.81 man00150m csb352h math1041 eengm4100 isys1002 08 6057cem mktg3504 mthm036 mtrx1701 mth3241 eeee3086 cmp-7038b cmp-7000a ints4010 econ2151 infs5710 fins5516 fin3309 fins5510 gsoe9340 math2007 math2036 soee5010 mark3088 infs3605 elec9714 comp2271 ma214 comp2211 infs3604 600426 sit254 acct3091 bbt405 msin0116 com107/com113 mark5826 sit120 comp9021 eco2101 eeen40700 cs253 ece3114 ecmm447 chns3000 math377 itd102 comp9444 comp(2041|9044) econ0060 econ7230 mgt001371 ecs-323 cs6250 mgdi60012 mdia2012 comm221001 comm5000 ma1008 engl642 econ241 com333 math367 mis201 nbs-7041x meek16104 econ2003 comm1190 mbas902 comp-1027 dpst1091 comp7315 eppd1033 m06 ee3025 msci231 bb113/bbs1063 fc709 comp3425 comp9417 econ42915 cb9101 math1102e chme0017 fc307 mkt60104 5522usst litr1-uc6201.200 ee1102 cosc2803 math39512 omp9727 int2067/int5051 bsb151 mgt253 fc021 babs2202 mis2002s phya21 18-213 cege0012 mdia1002 math38032 mech5125 07 cisc102 mgx3110 cs240 11175 fin3020s eco3420 ictten622 comp9727 cpt111 de114102d mgm320h5s bafi1019 math21112 efim20036 mn-3503 fins5568 110.807 bcpm000028 info6030 bma0092 bcpm0054 math20212 ce335 cs365 cenv6141 ftec5580 math2010 ec3450 comm1170 ecmt1010 csci-ua.0480-003 econ12-200 ib3960 ectb60h3f cs247—assignment tk3163 ics3u ib3j80 comp20008 comp9334 eppd1063 acct2343 cct109 isys1055/3412 math350-real math2014 eec180 stat141b econ2101 msinm014/msing014/msing014b fit2004 comp643 bu1002 cm2030
联系我们
EMail: 99515681@qq.com
QQ: 99515681
留学生作业帮-留学生的知心伴侣!
工作时间:08:00-21:00
python代写
微信客服:codinghelp
站长地图