CDS Candidate Guide 2020
CDS Candidate Guide 2020
CDS Candidate Guide 2020
TM
CANDIDATE
GUIDE
2020
W WW.ADAS CI.ORG
CDSTM Candidate Guide 2020
1
Content Introduction 3
Developed by the world’s leading data science practitioners, the designation
Reference Textbooks 12
CDS provides Reference Textbooks, practice exams and more to help you get ready
for the Exam. Due to the sizable amount of material covered
2
CDSTM Candidate Guide 2020
Introduction The Chartered Data Scientist (CDS™) has set the global standard for data
science. Developed by the world’s leading data science practitioners, the designation
signifies a mastery of the skills and knowledge needed to help organizations succeed in
today’s rapidly changing landscape. Its curriculum is updated annually by a group of
distinguished professionals and leading academics of diverse backgrounds, ensuring that
the designation meets the evolving demands of the global industry. By achieving your
-
CDS certification, you indicate to potential employers that you are requisite knowledge
of the field, providing you with an edge in your career and professional development.
3
CDSTM Candidate Guide 2020
BECOMING A The first step to becoming an CDS is passing a rigorous exam. To enroll in the program,
CERTIFIED CDS register for the CDS Exam online via the ADASCI website. CDS Exam is offered
throughout the year. After achieving a passing score on CDS Exam, candidates must
demonstrate a minimum of two years of full-time work experience in data science or a
related field to complete their certification. If you have questions about whether your
work experience qualifies, please contact info@adasci.org
CDS EXAM
THREE HOURS/150 QUESTIONS
4
CDSTM Candidate Guide 2020
CAREER CHANGERS
Whether you work in data science or are interested in transitioning to a data science role,
becoming an CDS can help accelerate your career. Professionals from non- data science
roles become CDS in order to develop specialized, practical knowledge that can be
applied in the industry. Undertaking the rigorous course of study to become an CDS
signals a commitment to an area of technology that is growing rapidly across the globe.
STUDENTS
Students with an interest in data science may elect to sit for the CDS Exam during or
immediately after completing their studies. The CDS curriculum can complement their
prior coursework or help them develop a foundation of specialized knowledge that goes
beyond their academic curriculum. Since the CDS Exam is practitioner- driven, earning
the CDS designation demonstrates to future employers that they are able to master
complex real-world challenges.
5
CDSTM Candidate Guide 2020
65
CDSTM Candidate Guide 2020
The CDS certification is by far the best known and most respected designation for data
science. As an CDS, you’ll have a competitive advantage that can help you stand out to
employers. By earning your certification, you’ll demonstrate that you possess the
knowledge and tools necessary to assess and manage the challenges associated with the
dynamic financial services industry.
7
CDSTM Candidate Guide 2020
Exam Development,
Structure and Content
EXAM DEVELOPMENT
The CDS Program is developed under the effectively contribute to their
guidance of the ADASCI Committee, organizations.
which is comprised of prominent global
data science experts and leaders. The EXAM STRUCTURE
CDS Committee establishes the topic The CDS consists of one computer based,
areas tested on the Exam on an annual multiple-choice exams. The CDS Exam
basis. To further align with industry - consists of 150 equally weighted
needs and calibrate our understanding of questions. Candidates are allotted three
the demands of the global risk hours to complete the Exam. The Exams
management community, we also are offered in person in American English
conduct formal surveys designed to throughout the year. The Exams are
determine the knowledge, skills and comprehensive, practice-oriented
abilities required of effective data assessments that cover the fundamental
scientists. This process helps ensure that tools and techniques used in data science,
successful candidates are prepared to their underlying theories.
8
CDSTM Candidate Guide 2020
Section 1:
Probability Theory, Statistics and Linear independence, Central limit theorem,
Algebra (12%) Subtopics: Counting, Frequentist significance tests and
Random variables, distributions, confidence intervals, Maximum
quantiles, mean-variance, p- Value, Likelihood Estimation, Bayes’ theorem
Confidence Interval, Hypothesis testing, and Bayesian statistics, Scalars, Vectors,
t-test, z-test, Chi Square test, Analysis of Matrices, and Tensors. Multiplying
Variance (ANOVA), Conditional Matrices and Vectors, Eigen
probability, base rate fallacy, Joint decomposition, Singular Value
distributions, covariance, correlation, Decomposition.
Section 2:
Data Engineering and Databases (8%) processing, data management, data
Subtopics: Relational databases, access, governance and integration,
Non-relational databases, key-value operations and security, SQL.
stores, batch processing, in-memory
Section 3:
Exploratory Data Analysis (8%) Scatter Plot, Contour Plot, Histogram,
Subtopics: Data Visualization, Box Plot, Bar Chart, Line Chart.
Section 4:
Supervised Learning and Unsupervised Support vector machine (SVM) and
Learning (15%) Subtopics: Linear and kernels, Model selection and model
Non-linear Models, Classification, selection criteria, ensemble learning -
Regression, K-Nearest Neighbours, bagging and boosting, expectation
Naïve Bayes, Clustering, K-Means maximization (EM) algorithm, Hidden
Clustering, Hierarchical Clustering, Markov models, Bayesian networks,
Perceptron learning rule, various learning Probabilistic inference. Association Rule
errors, regularization, estimator bias- Learning, Reinforcement Learning,
variance trade-off, active learning, Time-Series Analysis, Cross-Validation.
9
CDS
By AIM
TM
Candidate
Research &Guide 2020
igsaw Academy
Section 5:
Neural Networks and Deep Learning Neural Networks, Representation
(11%) Subtopics: Feedforward Learning, Autoencoders, Deep
Networks, Backpropagation Learning, Generative Models, Factor Analysis,
Gradient Descent, Regularization Principal Component Analysis,
techniques, Optimization techniques for Independent Component Analysis,
neural networks, Convolutional t-Distributed Stochastic Neighbour
Networks, Recurrent and Recursive Embedding (t-SNE)
Section 6:
Natural Language Processing (8%) Information Retrieval, Parsing, Part of
Subtopics: Text Classification, Language Speech (POS) Tagging, Sequence
Modelling, Sentiment Analysis, modelling.
Section 7:
Computer Vision (8%) Subtopics: Image Recognition, Image Tagging, Video
Processing, Image Classification, Object Analytics.
Section 8:
Deployment and Model management reporting and visualization mechanisms
(8%) Subtopics: Deployment of machine for model performance.
learning model, tracking model quality,
Section 9:
Python and R (10%) Subtopics: List, H2O, Data Analysis, Data Visualization,
Dictionary, Array, Conditional state- Implementation of Machine Learning
ments, Loops, Data Frame, Function, Algorithms.
File, Sci-Kit Learn, Keras, TensorFlow,
Section 10:
PBusiness and data science (12%) stakeholders, understanding constraints
Subtopics: Identify stakeholders, and scope of data science projects,
Handling data privacy concerns, Defining and communicating business
determining problem-data science fit, benefits, identifying data sources and
defining problem statement for multiple creating initial reports, Decision
Modelling.
10
CDSTM Candidate Guide 2020
Payments
and Fees
11
CDSTM Candidate Guide 2020
Reference
Textbooks
CDS provides Reference Textbooks, practice exams and more to help you get ready for
the Exam. Due to the sizable amount of material covered, we suggest that you use a
weekly study schedule. Preparation time will vary based on your prior professional
experience, academic background and familiarity with the curriculum’s concepts.
Preparing for the Exam at the last minute is strongly discouraged.
REFERENCE TEXTBOOKS:
Section 1:
A Course in Probability Theory, Kai Lai Robert Tibshirani, and Trevor Hastie,
Chung, Academic Press. An Introduction Springer Publication. Introduction to
to Statistical Learning: With Applications Probability Models, 9th Edition,
in R, Daniela Witten, Gareth James, Sheldon M.
Section 2:
Database System Concepts Textbook by The Big Ideas Behind Reliable, Scalable,
Avi Silberschatz, Henry F. Korth, and S. and Maintainable Systems, Martin
Sudarshan, McGraw Hill Publication. Kleppmann, O’Reilly Publication.
Designing Data-Intensive Applications:
12
CDSTM Candidate Guide 2020
Section 3:
Practical Statistics for Data Scientists: 50 Andrew Bruce, O’Reilly Publication.
Essential Concepts, Peter Bruce and
Section 4:
Pattern Recognition and Machine Mining, Inference and Prediction, 2nd
Learning, Christopher Bishop, Springer Edition, T Hastie, R Tibshirani and J
Publication. Machine Learning, Tom M. Friedman, Springer Series in Statistics,
Mitchell, McGraw Hill Publication. The Springer Publications.
Elements of Statistical Learning: Data
Section 5:
Deep Learning Book by Aaron C. Murphy, MIT Press. Neural Networks
Courville, Ian Goodfellow, and Yoshua and Learning Machines, 3rd Edition,
Bengio, MIT Press. Machine Learning A Simon Haykin, Pearson Publication.
Probabilistic Perspective, Kevin P.
Section 6:
Foundations of Statistical Natural Press. Natural Language Processing with
Language Processing, Christopher D. Python, Steven Bird, Ewan Klein and
Manning and Hinrich Schutze, The MIT Edward Loper, O’Reilly Publication.
Section 7:
Computer Vision: Algorithms and Applications, Richard Szeliski, Springer
Publication.
13
CDSTM Candidate Guide 2020
Section 8:
Evaluating Machine Learning Models, Applications, Emmanuel Ameisen,
Alice Zheng, O’Reilly Publication. O’Reilly Publication.
Building Machine Learning Powered
Section 9:
Python Cookbook: Recipes for Techniques to Build Intelligent Systems,
Mastering Python 3, 3rd Edition, David 2nd Edition, Aurelien Geron, O’Reilly
Beazley & Brian K. Jones, O’Reilly Publication. R for Data Science: Import,
Publication. Hands-On Machine TIDY, Transform, Visualize, and Model
Learning with Scikit-Learn, Keras and Data, Hadley Wickham and Garrett
TensorFlow: Concepts, Tools and Grolemund, O’Reilly Publication.
Section 10:
Laursen GHN, Thorlund J (2016) Hoboken, NJ). Business Analytics, 2nd
Business Analytics for Managers: Taking Edition, James Evans, Pearson
Business Intelligence Beyond Reporting, Publication.
2nd ed. (John Wiley & Sons,
14
CDSTM Candidate Guide 2020
Before proceeding for the Chartered Data Scientist (CDS) exam, the candidates are
required to accept and adhere to the following terms and conditions and privacy policy
agreements.
1. Important Note
1.1 These terms and conditions are the registration.
entered between the Candidate (You as a
prospective awardee of Chartered Data 1.2 The ADaSci has the right to change,
Scientist™) and the ADaSci (Association and/or update these terms and conditions
of Data Scientists) as on the date when from time to time.
the candidate has opted to proceed with
15
CDSTM Candidate Guide 2020
3.Eligibility Criteria
3.1 The minimum age of the candidate to years.
appear for the CDS exam must be 18
years on the date of registration. 3.4 The candidate must have at least two
years of relevant experience as a Data
3.2 If the age of the candidate is between Scientist for the award of this charter.
13 to 18 years, there must be an However, the candidate can appear in
undertaking signed by the parents or the the exam and receive the result, but the
guardian. award of Chartered Data Scientist will be
put on hold until the candidate attains
3.3 The age must not be less than 13 the two years of experience.
16
CDSTM Candidate Guide 2020
7.Award of Charter
7.1 The award of Chartered Data including experience details provided by
Scientist will be provided to the candi- the candidate during registration.
date by ADaSci on successfully passing
the exam along with the following 7.1.2 If the candidate has not attained
conditions: the two years of experience at the time of
exam and when he or she provides the
7.1.1 Verification of all the information required experience details later.
8.Re-Evaluation
of the exam.
8.1 As per the current regulations, there
is not any provision for the re-evaluation
17
CDSTM Candidate Guide 2020
10.Validity
10.1 The awarded charter will have the life-time validity.
11.Attempts
11.1 An individual can take any number the year and there is not any limit on
of attempts to pass this exam throughout attempts.
13.Other Notes
13.1 The ADaSci does not provide the or providing any type of job assistance.
following guarantee to the owner of this
charter: 13.1.2 To get admission to any institu-
tion for study.
13.1.1 To get a job in any organization
18
CDSTM Candidate Guide 2020
19
#280, 2ND FLOOR, 5TH MAIN, 15TH A CROSS
RD, SECTOR 6, HSR LAYOUT, BENGALURU,
KARNATAKA 560102
WWW.ADASCI.ORG