Data Analysis and Management Concepts Quiz

Data that can be expressed as numbers, such as ratios, percentages, or counts.

Data that describes qualities or characteristics, such as opinions, feelings, or descriptions.

Data that has distinct separate values, with gaps between them. It is countable.

Data that can take on any value within a range, with no gaps between values. It is measurable.

A database is a structured collection of information, organized and stored electronically.

A data source is the origin of the data being used. It can be the initial point of data creation or a refined version used by other processes.

A database management system (DBMS) is software used to manage and control a database.

Data that represents categories that can be ordered or ranked, but the distance between categories is not necessarily equal. Examples include education level (Elementary, Middle, High School, College) or job position (Manager, Supervisor, Employee).

Statistical tests used to compare the frequency or proportion of observations in different categories. They are often used for analyzing ordinal and nominal data.

Non-parametric tests used to analyze ordinal data. They do not assume a specific distribution of data and are useful for comparing groups when the data is ranked.

A non-parametric test used to compare two related groups of ordinal data, such as before-and-after measurements. It assesses whether there is a significant difference in ranks between the groups.

A non-parametric test used to compare two independent groups of ordinal data. It assesses whether there is a significant difference in the distribution of ranks between the groups.

The process of combining multiple datasets into one larger dataset. This step is crucial for consolidating information, reducing redundancy, and improving the completeness of the data.

The process of identifying and correcting errors, inconsistencies, and inaccuracies within a dataset. Data cleaning ensures that the data is accurate, consistent, and reliable, leading to better insights and model performance.

The systematic identification and removal of redundant or irrelevant observations from a dataset. This step involves scrutinizing data entries for duplicate records, irrelevant information, or data points that do not contribute meaningfully to the analysis.

The process of addressing structural issues in a dataset, such as inconsistencies in data formats, naming conventions, or variable types. It involves standardizing formats, correcting naming discrepancies, and ensuring uniformity in data representation.

Incorrect or inconsistent data can negatively impact the performance of a machine learning model, leading to inaccurate predictions and misleading insights.

Data cleaning plays a significant role in building a machine learning model, as it ensures the data used for training is accurate, reliable, and free of errors.

The quality of data is often more important than the complexity of the algorithms used in machine learning.

Raw data is often noisy, incomplete, and inconsistent, which makes it unreliable for analysis and model building.

Data points significantly deviating from the norm in a dataset. They can distort analysis and impact insights.

A technique to minimize the impact of outliers on data analysis by either removing them or transforming their values.

The challenge of dealing with missing values in your dataset, which can lead to inaccurate analysis.

Missing values where there's no pattern or reason why data is absent. It's like a random missing piece of a puzzle.

Missing values where the reason for missing data depends on other data in the dataset. Think of a missing exam score due to low grades.

Missing values where the reason they're missing is related to the value itself. For example, high-income individuals might not disclose their income.

A common strategy for handling missing data that involves removing entire rows with missing values from the dataset. This method is simple but might sacrifice valuable data.

A strategy for handling missing data that replaces missing values with estimated or imputed values based on statistical methods.

A technique that helps in handling skewed data by transforming the distribution to resemble a normal distribution. This reduces the impact of outliers on the data, making models more robust.

A process of grouping data points into bins or intervals. It's used to handle noisy data and can reduce overfitting in machine learning models.

Splitting a feature into multiple parts to create new features. This helps algorithms better understand patterns and improve model performance.

A technique for converting categorical data into a format that can be easily understood by machine learning algorithms. It assigns a unique numerical value to each category, representing them in binary form.

Features derived from previous time series values. They help capture patterns, such as seasonality and trends, for better time series forecasting.

Features calculated using a predefined window of data points. They help capture local variations and trends.

Features capturing specific points in time, such as day of the week, month, or holiday. They help model time-dependent patterns.

A technique for handling categorical data that is typically used when the categories are ordinal (have an order). Each category is assigned a numerical value, maintaining the order from the original data.

The percentage of missing values in a column, calculated by dividing the number of missing values by the total number of observations.

A threshold value that determines whether a variable (column) with a high percentage of missing values should be removed. The variable is dropped if its ratio exceeds the threshold.

A type of feature selection method that combines the benefits of filter and wrapper methods. These methods are faster than wrappers but generally provide more accurate results than filters. They iterate through the data to identify important features.

L1 and L2 are two popular techniques for preventing overfitting by adding penalties to the model's parameters. This approach effectively shrinks less relevant features by forcing their coefficients towards zero.

A feature selection technique that uses a Random Forest model (an ensemble of decision trees) to evaluate the importance of each feature based on its contribution to the model's accuracy.

A process of combining data from multiple tables (datasets) based on a defined relationship between their columns, typically using a shared column (key) for alignment.

A type of data merge that combines rows from two tables based on matching values in a shared column, ensuring that only rows with matching values in the shared columns from both tables are included in the result.

A type of data merge that combines all rows from the left table with matching rows from the right table. If there is no match in the right table, NULL values are filled in for columns from the right table.