AP Comp Sci Unit 5 Study Guide PDF

Vocab Machine Learning Definition: Machine learning is a type of computer programming where computers learn from data without being explicitly programmed. They use patterns in data to make decisions or predictions. Example: A program that learns to recognize cats in photos by loo...

Vocab Machine Learning Definition: Machine learning is a type of computer programming where computers learn from data without being explicitly programmed. They use patterns in data to make decisions or predictions. Example: A program that learns to recognize cats in photos by looking at thousands of images labeled as "cat" or "not cat." Cleaning Data Definition: Cleaning data means fixing mistakes in the data, like removing duplicates, correcting errors, and filling in missing values, to make it accurate and ready for analysis. Example: Fixing a list of ages where some values are typed as "twenty" instead of "20." Histogram Definition: A histogram is a type of graph that shows how often different ranges of values appear in a dataset. Example: A histogram of student test scores might show how many students scored between 0-10, 11-20, 21-30, etc. Data Analysis Definition: Data analysis is the process of examining and interpreting data to find patterns, trends, or insights that can help answer questions or solve problems. Example: Analyzing sales data to understand which products are selling the most. Training Data Definition: Training data is the data used to teach a machine learning model how to make predictions or decisions. Example: A machine learning model trained with pictures of dogs and cats, labeled "dog" or "cat," to learn how to recognize animals in new photos. Filtering Data Definition: Filtering data means selecting only specific parts of a dataset based on certain criteria, like choosing data from a particular time period or range. Example: Filtering a list of students to only show those who scored above 90 on a test. Bar Chart Definition: A bar chart is a graph that uses bars to represent different categories of data, with the length of each bar showing how much or how many there are in that category. Example: A bar chart showing the number of students in different grade levels, with one bar for each grade. Big Data Definition: Big data refers to extremely large datasets that are too complex for traditional data processing methods to handle easily. Example: The amount of data generated by social media platforms, like Facebook or Twitter, every day. Algorithm Definition: An algorithm is a set of instructions or steps that tell a computer how to perform a task or solve a problem. Example: An algorithm to sort a list of names alphabetically. Correlation Definition: Correlation refers to the relationship or connection between two sets of data. If one changes, the other might change as well. Example: There is often a positive correlation between studying more and getting better test scores. Crowdsourcing Definition: Crowdsourcing is when a large group of people (usually online) contribute to a project or help solve a problem. Example: Using volunteers online to label photos for a machine learning project. Metadata Definition: Metadata is data that describes other data. It gives information about the data, like when it was created, who created it, or how it was collected. Example: The metadata of a photo might include the date it was taken, the camera settings, and the location. Data Bias Definition: Data bias happens when data is not representative of the entire population or is skewed in a way that leads to unfair or incorrect conclusions. Example: If a survey about video game preferences only includes responses from young people, the results may be biased toward their interests. Information Definition: Information is processed data that has meaning or is useful for making decisions or understanding something. Example: A report showing the total sales for each product over a month is information because it helps understand sales trends. Open Data Definition: Open data is data that is freely available to the public, meaning anyone can access, use, or share it. Example: Government data about traffic patterns that is made available online for anyone to use. Scatter Plot Definition: A scatter plot is a graph that shows data points on a two-dimensional grid. Each point represents one pair of values, and it helps to see if there’s a relationship between those values. Example: A scatter plot showing the relationship between hours studied and test scores, where each point represents a student's study time and score. Citizen Science Definition: Citizen science is when regular people (not scientists) participate in collecting or analyzing data for scientific research. Example: People recording bird sightings and sending their data to researchers studying migration patterns. Cross Tab Chart Definition: A cross tab chart (or cross-tabulation) is a table that displays the relationship between two or more variables in rows and columns, often used to summarize and compare data. Example: A table showing how different age groups (rows) prefer different types of music (columns). Key Topics 1. Extracting Information from Data Definition: The process of analyzing raw data to uncover useful patterns, trends, or insights. Methods: ○ Data Collection: Gather data from sources like surveys, sensors, databases, etc. ○ Data Cleaning: Ensure data is accurate, consistent, and free of errors. ○ Data Analysis: Use algorithms or statistical methods to identify patterns or insights from data. 2. Using Programs with Data Definition: Writing programs to process data, manipulate it, and extract insights automatically. Examples: ○ Sorting data ○ Filtering data based on certain criteria ○ Creating algorithms to analyze trends or predict future events (e.g., machine learning algorithms) 3. Computing Bias Definition: The systematic favoritism or skewing of results due to flawed data, biased algorithms, or improper sampling. Types of Bias: ○ Sampling Bias: Data collected in a way that is not representative of the entire population. ○ Algorithmic Bias: Algorithms that produce results based on biased input data, leading to unfair outcomes. ○ Data Bias: When data contains pre-existing biases, such as historical inequalities or prejudice. 4. Crowdsourcing Definition: The process of obtaining data, services, or solutions by soliciting contributions from a large group of people, typically through the internet. Examples: ○ Crowdsourced data for scientific research, such as through citizen science projects. ○ Crowdsourced solutions for machine learning tasks, like labeling images. Data Visualizations 1. Bar Graph Definition: A graph that uses rectangular bars to represent data. The length of each bar corresponds to the value it represents. When to Use: ○ Comparing discrete categories. ○ Example: Comparing the sales of different products. 2. Histogram  Definition: A type of bar graph used to represent the distribution of numerical data. The x-axis represents ranges of values, and the y-axis represents the frequency of data within those ranges. When to Use: ○ Understanding the distribution of a data set (e.g., how the ages of a group of people are spread out). ○ Example: A histogram of test scores showing how many students fall into different score ranges. 3. Cross Tab Chart (Contingency Table) Definition: A table used to display the relationship between two categorical variables. When to Use: ○ Analyzing how different categories interact with each other. ○ Example: A table showing how the preferences of different age groups relate to product choices. 4. Scatter Plot Definition: A graph that uses dots to represent values for two different variables. Each dot represents a data point with values for both variables plotted along the x- and y-axes. When to Use: ○ Identifying relationships between two variables, such as whether they are correlated. ○ Example: A scatter plot of hours studied versus test scores. Practice Questions: 1. Extracting Information from Data: ○ You are given a dataset of student grades in a class. What steps would you take to analyze this data and extract insights about the overall performance of the class? 2. Machine Learning: ○ Explain how a machine learning algorithm might be used to predict whether a customer will buy a product based on their previous purchasing history. 3. Bias in Data: ○ How can sampling bias impact the accuracy of data collected in a survey about public opinion on a policy? 4. Crowdsourcing:  ○ What is one example of crowdsourcing in the field of data science, and how does it help improve data collection? 1. Extracting Information from Data Definition: Extracting information from data refers to analyzing and interpreting data to uncover useful insights or patterns. Techniques: ○ Data Collection: Gathering data through various methods, such as surveys, sensors, or logs. ○ Data Cleaning: Removing or correcting errors in data, such as duplicates or inconsistencies. ○ Data Transformation: Changing the format or structure of data for easier analysis (e.g., converting data to CSV format). ○ Data Filtering: Selecting specific subsets of data to focus on (e.g., filtering for a certain time range or category). ○ Aggregation: Combining data from multiple sources or summarizing it (e.g., calculating the average of a data set). ○ Visualization: Presenting data in graphical formats like charts, graphs, or tables to make it easier to understand and interpret. Example: ○ You have a dataset of students' test scores. To extract useful information, you might compute the average score, identify trends, or filter out scores below a certain threshold. 2. Using Programs with Data Definition: Programs can be used to analyze, manipulate, and visualize data efficiently. Tools & Concepts: ○ Algorithms: Steps or processes used to solve a problem or perform a task. Algorithms can be used to process large datasets (e.g., sorting or searching data). ○ Data Structures: Ways to organize data in programs, such as lists, arrays, dictionaries, and tables. ○ Programming Languages: You might use languages like Python, JavaScript, or SQL to process and analyze data. ○ APIs (Application Programming Interfaces): APIs can be used to retrieve data from external sources (e.g., social media data, weather data) and integrate it into programs. Example:  ○ Writing a Python program that processes a dataset of temperatures to calculate the maximum, minimum, and average temperature. 3. Computing Bias in Data Definition: Bias in data occurs when data is not representative of the entire population or is skewed in a certain direction. Types of Bias: ○ Sampling Bias: When the sample of data collected does not represent the entire population (e.g., only surveying people from one geographic location). ○ Measurement Bias: When data is measured inaccurately or inconsistently. ○ Confirmation Bias: When data is interpreted or analyzed in a way that confirms preexisting beliefs or expectations. ○ Algorithmic Bias: When algorithms used to analyze or make decisions from data unintentionally favor certain groups or outcomes. Example: ○ If an algorithm that predicts job performance is trained on historical data from a specific group (e.g., a particular gender or ethnicity), it may inadvertently produce biased results that favor that group. 4. Crowdsourcing Definition: Crowdsourcing involves obtaining input, data, or services from a large group of people, typically from an online community. Types of Crowdsourcing: ○ Crowd Labor: Assigning small tasks to a large group of people to complete (e.g., labeling images for machine learning). ○ Crowd Wisdom: Harnessing the collective knowledge or opinions of a large group to solve problems or make decisions (e.g., using public votes to determine the best product idea). ○ Crowdfunding: Gathering funds from a large number of people, typically via online platforms. Advantages: ○ Diverse Perspectives: Crowdsourcing allows a wide range of ideas and inputs from people with different backgrounds and expertise. ○ Efficiency: Tasks can be completed more quickly because many people can contribute simultaneously. ○ Cost-Effectiveness: It can be less expensive than hiring specialists or using traditional methods. Disadvantages:  ○ Quality Control: Ensuring the accuracy or quality of contributions can be challenging. ○ Potential for Bias: Crowdsourcing may amplify certain biases or skew results if the crowd is not diverse enough. Example: ○ Wikipedia uses crowdsourcing to gather and edit its articles. Anyone can contribute to the content, but there are moderators to help ensure accuracy and quality. 6. Data Interpretation and Communication Definition: Interpreting data involves making sense of the data analysis results and presenting them in a clear and understandable way. Tools: ○ Charts and Graphs: Bar charts, pie charts, line graphs, scatter plots, etc., are commonly used to present data visually. ○ Summary Statistics: Measures such as mean, median, mode, range, and standard deviation are often used to summarize and interpret data. ○ Narrative: Telling a story based on the data. For instance, explaining what the data reveals about trends, behaviors, or outcomes. Example: ○ After analyzing survey results, you might create a pie chart to show how respondents feel about a certain issue and then write a summary of what those results mean.