QUALIFI Level 3 Diploma in Data Science

This unit provides learners with an overview of the evolution of data science, starting from the emergence of artificial intelligence and machine learning in the late 1950s, leading to the onset of the "big data" era in the early 2000s. It further explores contemporary applications of AI, machine learning, deep learning, and associated challenges.

Reference No : H/650/4951

This unit introduces learners to fundamental charts and visualisations, including creation and interpretation methods. It begins by elucidating visualisations' significance in comprehending data and distinguishing between effective and ineffective visual representations. The unit introduces learners to various basic chart and plot types, clarifying their intended use, interpretation guidelines, and situations where they are most and least appropriate. Subsequently, the unit zeroes in on the technology for generating charts and visualisations in Python, encompassing tools like Seaborn, Matplotlib, and other Python libraries.

Reference No : K/650/4953

This unit is designed to give learners a foundational understanding of descriptive statistics, pivotal in data analysis and science. It covers various types of data and introduces key descriptive statistics, including measures of central tendency, multiple indicators of data spread (such as range, percentiles, variance, and standard deviation), measures of symmetry (skewness and kurtosis), and actions of co-variation (correlation and covariance). The unit also elucidates which descriptive statistics apply to data measured on different scales. Throughout this unit, learners will actively engage in the hands-on practice of manually calculating descriptive statistics for small datasets.

Reference No : L/650/4954

This unit introduces the fundamental concepts of data analytics. It aims to enable learners to distinguish between the roles of a Data Analyst, Data Scientist, and Data Engineer. Additionally, learners will gain an understanding of the data ecosystem, which encompasses databases and data warehouses, and become acquainted with key vendors within this ecosystem, along with exploration of various tools. The unit also covers essential tasks and processes in the data discovery phase, including data cleaning, approaches to address data quality issues, and methods for standardising data in preparation for analysis.

Reference No : M/650/4955

In this unit, learners are introduced to foundational data analysis using Python. Core concepts like Pandas DataFrames and Series and techniques for merging and joining data are covered. The unit further extends previous modules by instructing on data importing, employing Python for generating descriptive statistics for analysis and interpretation. Additionally, learners will acquire skills in utilising Python to enhance data quality and standardise data, particularly in preparing it for machine learning models.

Reference No : R/650/4956

This unit offers a broad overview (as opposed to an in-depth examination) of the three primary categories of machine learning: supervised, unsupervised, and reinforcement learning. It delves into the practical applications and real-world issues each method can address. Additionally, the unit provides a concise summary of each approach's distinctive features and potential challenges.

Reference No : T/650/4957

This unit introduces the multifaceted steps and procedures in constructing and assessing machine learning models. It explains the fundamental elements of the machine learning process, from data preparation and choosing the appropriate machine learning algorithm to partitioning data into training, testing, and validation sets to mitigate the underfitting risks. The unit also addresses identifying and resolving class imbalance, emphasising when such strategies are necessary. Given that many encountered machine learning models are supervised classification models, the unit acquaints learners with common performance metrics and how to interpret them. Lastly, the unit briefly discusses methods for managing model bias and variance

Reference No : Y/650/4958

This unit provides a foundational understanding of simple linear regression models, which are essential for predicting the value of one continuous variable based on another. Learners will gain the ability to estimate the line of best fit by computing regression parameters and comprehend the accuracy of this line. Furthermore, the unit expands on simple linear regression by introducing multiple and polynomial regression models, enabling the examination of relationships involving numerous variables. It outlines constructing simple, multiple, and polynomial linear regression models using Python, leveraging libraries like sci-kit-learn.

Reference No : A/650/4959

This unit serves as an introduction to logistic regression and its role as a classification algorithm. It delves into the fundamentals of binary logistic regression, including the logistic function, Odds ratio, and Logit function. Additionally, the unit clarifies the distinctions between linear and logistic regression. Learners will acquire the skills to construct and visually represent a logistic regression model using Python. The unit will also educate learners on when to opt for logistic regression over linear regression, how to accurately interpret the outcomes of logistic regression, and how to select the optimal logistic model that effectively characterises the relationship under examination.

Reference No : H/650/4960

This unit introduces the fundamental theory and practical application of decision trees. It elucidates the construction of basic classification trees by utilising the standard ID3 decision-tree construction algorithm. The unit further details how nodes are divided based on concepts from information theory, such as Entropy and Information Gain. Additionally, learners will gain hands-on experience constructing and evaluating decision tree models using Python.

Reference No : J/650/4961

This unit introduces an unsupervised machine learning algorithm: k-means clustering. It aims to impart learners with an understanding of the underlying principles of the k-means clustering algorithm, including methods for determining the optimal number of clusters. Additionally, learners will gain practical experience constructing and assessing k-means models using Python and acquire skills in visualising the resultant sets.

Reference No : K/650/4962

This unit introduces the emerging field of data science - synthetic data and its role in enhancing data privacy and security. In the contemporary information age, data collected by entities like Google and Facebook and giving these datasets is paramount. Inadvertent disclosure or leakage of this data poses a significant dual privacy. The unit covers topics on confidentiality, including data privacy, the imperative for privacy, and the legal framework surrounding it. It delves into conventional methods of safeguarding data privacy, such as anonymisation. Then, it introduces learners to differential privacy and mental challenges in balancing data privacy and data utility.

Reference No : L/650/4963

This unit introduces learners to another burgeoning field in data science - graphs and graph data science. It provides a beginner-friendly introduction to graph theory, a foundational concept underlying modern graph databases and analytics. The unit also explores the graph ecosystem, introducing concepts like Knowledge Graphs, Labelled Property Graphs, and RDF graphs for data storage and processing. Additionally, learners will be introduced to graph algorithms, essential tools for modelling, storing, retrieving, and analysing graph-structured data.

Reference No : M/650/4964