Schlussprüfung — Fachinformatiker Daten- und Prozessanalyse (English)

• A) The complete automation of business processes without human intervention
• B) The standardized graphical representation of business processes for all stakeholders
• C) The pure documentation of existing processes without optimization possibilities
• D) The replacement of all other process modeling methods such as UML or flowcharts

Correct Answer: B. BPMN serves the standardized graphical representation of business processes that both technical and non-technical people can understand. Option A is incorrect because BPMN does not cause complete automation. Option C is incorrect because BPMN is not only for documentation but also for identifying optimization potential. Option D is incorrect because BPMN is a specialized method that does not replace others.

• A) A single activity or task within a process
• B) A condition or rule that controls the process flow
• C) A point in the process where something happens
• D) A participant role or an organizational unit

Correct Answer: D. A pool represents a participant role or an organizational unit and separates the entire process from other processes. Option A describes an activity, not a pool. Option B describes a gateway, not a pool. Option C describes an event, not a pool.

• A) Event
• B) Lane
• C) Gateway
• D) Pool

Correct Answer: C. A Gateway determines the flow of the process based on conditions, rules, or parallelism. An Event (Option A) represents a point where something happens. A Lane (Option B) divides a pool into horizontal strips. A Pool (Option D) represents a participant role or organizational unit.

• A) Drawing the main process flow with Start and End Event

• A) Cube Schema
• B) Snowflake Schema
• C) Galaxy Schema
• D) Flat File Schema

Correct Answer: B. The Snowflake Schema is an extension of the Star Schema with further normalized dimension tables, while the other options are not common data models for data warehouse systems.

• A) Dimension Table
• B) Attribute Table
• C) Fact Table
• D) Master Data Table

Correct Answer: C. The fact table contains the numeric measurements, while the other options contain descriptive data or other types of reference data.

• A) SCD Type 1 (Overwrite)
• B) SCD Type 2 (Historization)
• C) SCD Type 3 (Addition with historical value)
• D) SCD Type 4 (New Table)

Correct Answer: C. SCD Type 3 adds new columns to store historical values, while Type 1 overwrites old values and Type 2 creates a complete historization with time periods.

• A) Snowflake Schema
• B) Star Schema
• C) Normalized Schema
• D) Entity-Relationship Schema

Correct Answer: B. The Star Schema is characterized by a

• A) ETL processes data in the cloud, ELT locally
• B) In ETL, transformation occurs before loading, in ELT after loading
• C) ETL always uses staging areas, ELT does not
• D) ELT is only suitable for Big Data environments

Correct Answer: B. The essential difference lies in the timing of the transformation: ETL transforms data before loading into the target system, while ELT loads data first and then transforms it in the target system.

• A) Completeness
• B) Consistency
• C) Accuracy
• D) Validity

Correct Answer: B. Consistency ensures that data matches across different systems or datasets. Completeness refers to the presence of all expected data, accuracy to the correctness of values, and validity is a broader term for compliance with established rules.

• A) Pandas
• B) NumPy
• C) Great Expectations
• D) SQLAlchemy

Correct Answer: C. Great Expectations is the framework presented in the module for automated monitoring of data quality. Pandas and NumPy are libraries for data manipulation and numerical calculations, and SQLAlchemy is a toolkit for SQL databases.

• A) Data Cleansing
• B) Data Profiling
• C) Data Modeling
• D) Data Aggregation

Correct Answer: B. Data profiling is the systematic process for examining the characteristics of data holdings. Data cleansing refers to the removal of errors, data modeling to structure definition, and data aggregation to the summarization of data.

• A) TensorFlow
• B) Matplotlib

• A) Window Functions can only be applied to numeric data
• B) Window Functions do not group rows but perform calculations on a window of rows
• C) Window Functions always require a GROUP BY clause
• D) Window Functions can only be used with the DISTINCT clause

Correct Answer: B. Window Functions operate on a window of rows without grouping them, while regular aggregate functions group rows and return one value per group. Option A is incorrect because Window Functions work on various data types. Option C is incorrect because Window Functions work without GROUP BY. Option D is incorrect because DISTINCT is not used with Window Functions.

• A) CTEs always improve query performance
• B) CTEs enable recursive queries
• C) CTEs increase the readability and modularity of queries
• D) CTEs can only be used with SELECT statements

Correct Answer: C. CTEs improve readability and modularity by breaking down complex queries into logical, named parts. Option A is incorrect because CTEs do not always improve performance. Option B is partially correct but not the main advantage. Option D is incorrect because CTEs can be used with INSERT, UPDATE, DELETE, etc.

• A) Compress data to save storage space
• B) Move data between different tables
• C) Change data structure by transforming rows into columns and vice versa
• D) Encrypt data to increase security

Correct Answer: C. Pivot/Unpivot operations change the data structure by transforming rows into columns (Pivot) or columns into rows (Unpivot), often for reports or dashboards. Option A is incorrect because it's not primarily about compression. Option B is incorrect because it's not about moving data. Option D is incorrect because it's not about encryption.

• A) The estimated costs of the query
• B) The exact records that will be returned by the query
• C) The used indexes
• D) The join methods employed

Correct Answer: B. An EXPLAIN plan shows how the database will execute the query, including estimated costs, indexes, and join methods, but it does not show the actual data that will be returned. Option A is incorrect because estimated costs are shown. Option C is incorrect because used indexes are shown. Option D is incorrect because join methods are shown.

• A) DAX can only work with numbers, Excel formulas also with text
• B) DAX is optimized for complex data models, Excel formulas for simple spreadsheet calculations
• C) DAX does not support cell references, only whole tables
• D) DAX can only be used in Power BI, Excel formulas are universally applicable

Correct Answer: B. DAX is specifically developed for BI tools and complex data models, while Excel formulas are optimized for tabular calculations. DAX can work with text and is also used in other Microsoft products.

• A) Maximum information density to show all relevant data at a glance
• B) Use as many different chart types as possible for variety
• C) Consistent visualization and focusing on the most important KPIs
• D) Avoid interaction options to reduce confusion

Correct Answer: C. Consistent visualization and focusing on the most important KPIs are crucial for clear, understandable dashboards. Too much information or chart types can overwhelm, while interaction improves the user experience.

• A) Metabase offers significantly more visualization options
• B) Metabase does not require SQL knowledge for basic operations
• C) Metabase is better suited for large enterprise environments
• D) Metabase provides more advanced DAX functions

Correct Answer: B. Metabase is designed to allow users without deep SQL knowledge to explore data and create dashboards, while Power BI requires more technical knowledge for similar tasks.

• A) A pure financial system for measuring profitability
• B) A strategic management system that divides corporate objectives into four perspectives
• C) A method for pure process optimization without strategic relevance
• D) A tool for pure data collection without analysis functions

Correct Answer: B. The Balanced Scorecard divides corporate objectives into four perspectives: Finance, Customers, Internal Processes, and Learning & Growth. Option A is incorrect because the Balanced Scorecard goes beyond finances. Option C is incorrect because it includes strategic goals. Option D is incorrect because it includes analysis functions.

• A) Simple, Measurable, Achievable, Relevant, Time-bound
• B) Specific, Measurable, Achievable, Relevant, Time-bound
• C) Strategic, Measurable, Actionable, Realistic, Timely
• D) Significant, Measurable, Applicable, Reliable, Targeted

Correct Answer: B. SMART-KPIs must be Specific, Measurable, Achievable, Relevant, and Time-bound. Option A is incorrect because "Simple" is not part of the acronym. Option C is incorrect because "Strategic" and "Timely" are not correct abbreviations. Option D is incorrect because the terms do not fit the SMART concept.

• A) The manual design of business processes
• B) The objective analysis of business processes based on event logs
• C) The development of new software for process management
• D) The creation of organizational charts

Correct Answer: B. Process Mining enables the objective analysis of business processes based on event logs to understand and optimize actual flows. Option A describes process design, Option C software development, and Option D organizational structure, but not the core function of Process Mining.

• A) A log of system errors and exceptions
• B) A structured dataset that captures the sequence of events in a process with timestamps, case IDs, and activity information
• C) A collection of user comments on process steps
• D) A report on employee performance

Correct Answer: B. An Event-Log is a structured dataset that captures the sequence of events in a process with timestamps, case IDs, and activity information and forms the basis for every process mining analysis. Option A describes error logs, Option C user feedback, and Option D performance evaluations, but not the specific structure of an Event-Log.

• A) The prediction of future process changes
• B) The verification of the alignment between the actual process and the target model
• C) The automation of process steps
• D) The visualization of process flows without prior knowledge

Correct Answer: B. Conformance Checking verifies the alignment between the actual process (as depicted in the event log) and the target model to uncover deviations, inefficiencies, and violations. Option A describes trend analyses, Option C process automation, and Option D discovery analyses, but not the core of Conformance Checking.

• A) The identification of the most common proc

• A) Variance
• B) Standard deviation
• C) Range
• D) Interquartile range

Correct Answer: B. The standard deviation is the square root of the variance and is in the same unit as the original data. The variance is in squared units, while range and interquartile range are in the same unit but measure different aspects of dispersion.

• A) Correlation is always linear, causality is nonlinear
• B) Correlation describes a relationship, causality describes a cause-effect relationship
• C) Correlation can only exist between numerical variables, causality also between categorical ones
• D) Correlation is always positive, causality can also be negative

Correct Answer: B. Correlation merely describes that two variables vary together, while causality means that a change in one variable directly leads to a change in the other. Correlation is a necessary but not sufficient condition for causality.

• A) Mean
• B) Mode
• C) Median
• D) Arithmetic mean

Correct Answer: C. The median is the middle value in ordered data and is not influenced by extreme values (outliers). The mean and arithmetic mean are identical and are strongly influenced by outliers, while the mode is also robust but not always uniquely defined.

• A) Calculate the test statistic
• B) Formulate the null and alternative hypotheses
• C) Choose the significance level
• D) Collect the data

Correct Answer: B. The first step in hypothesis testing is to clearly formulate the null hypothesis (which assumes no effect) and the alternative hypothesis (which represents the research claim). This must be done before collecting data or calculating statistics.

• A) Maximizing data precision
• B) Preventing the identification of individuals
• C) Reducing data volume
• D) Accelerating data processing

Correct answer: B. k-anonymity aims to ensure that individuals in a dataset cannot be identified by ensuring each person shares identical attributes with at least k-1 others. The other options do not describe the main goal of k-anonymity.

• A) Complete removal of all personal data
• B) Ensuring that adding or removing a single person does not significantly change analysis results
• C) Reversibility of anonymization upon request
• D) Guaranteeing 100% accuracy of data analysis

Correct answer: B. Differential Privacy is based on the mathematical principle that individual data points do not have a significant impact on the result. The other options do not describe the core concept of Differential Privacy.

• A) Processing of data by a controller without external help
• B) Processing of data by a third party on behalf of the controller with a contractual agreement
• C) Automated processing of data without human intervention
• D) Processing of data for advertising purposes without consent

Correct answer: B. Processing of data by third parties means that a data processor processes personal data on behalf of the controller, where a contractual agreement is made to ensure compliance with data protection regulations.

• A) Checking whether actual processes match target models
• B) Identification of unknown or implicit process models from event data
• C) Calculation of process key performance indicators for monitoring
• D) Automation of business processes

Correct Answer: B. Process-Discovery aims to identify unknown process models from event data without making predefined assumptions. Option A describes Conformance Checking, option C KPI dashboards, and option D is process automation.

• A) XML
• B) CSV
• C) XES
• D) JSON

Correct Answer: C. XES (eXtensible Event Stream) is the standard format for Event Logs in Process-Mining. While CSV and JSON can also be used, XES was specifically developed for Process-Mining analyses. XML is related but not the primary format.

• A) Process Discovery uses KPI dashboards, Conformance Checking does not
• B) Process Discovery identifies unknown processes, Conformance Checking checks compliance with target models
• C) Process Discovery requires Event Logs, Conformance Checking does not
• D) Process Discovery is for real data, Conformance Checking only for simulated data

Correct Answer: B. Process Discovery identifies unknown p

• A) Supervised learning always uses neural networks, unsupervised learning does not
• B) Supervised learning requires labeled data, unsupervised learning works with unlabeled data
• C) Supervised learning is always more accurate than unsupervised learning
• D) Supervised learning can only work with numerical data, unsupervised learning can also work with categorical data

Correct Answer: B. The key difference lies in the use of labeled data in supervised learning, while unsupervised learning works without predefined labels. Option A is incorrect as both learning forms include various algorithms. Option C is not generally valid as accuracy depends on the problem statement. Option D is incorrect as both learning forms can work with different data types.

• A) Classification
• B) Clustering
• C) Regression
• D) Principal Component Analysis

Correct Answer: C. Regression is the prediction of continuous values like prices. Classification would be incorrect as it categorizes data. Clustering is unsupervised learning and PCA is used for dimensionality reduction, not prediction.

• A) Underfitting
• B) Overfitting
• C) The Bias-Variance Dilemma
• D) The Problem of High Dimensionality

Correct Answer: B.<st

• A) Reducing data size for faster processing
• B) Transforming and selecting features to improve model performance
• C) Fully automating the data process
• D) Eliminating all categorical variables from the dataset

Correct Answer: B. Feature Engineering aims to improve ML model performance through targeted transformation and selection of features, while the other options only represent partial aspects or misinterpretations of this process.

• A) To increase the amount of training data available
• B) To ensure the model can be evaluated on unseen data
• C) To reduce computational requirements
• D) To allow for parallel processing of data

Correct Answer: B. The split ensures that the model's performance can be evaluated on data it has never seen during training, providing an unbiased assessment of how well it will perform in real-world scenarios.

• A) It reduces the time needed for model training
• B) It provides a more robust estimate of model performance
• C) It eliminates the need for a test set
• D) It automatically selects the best model

Correct Answer: B. Cross-Validation provides a more robust estimate of model performance by averaging results across multiple different splits of the data, reducing the impact of how the data is divided.

• A) Accuracy
• B) F1-Score
• C) Number of features
• D) Training time

Correct Answer: B. For imbalanced data, the F1-Score is more appropriate than Accuracy because it considers both Precision and Recall, providing a better measure of the model's performance on the minority class.