Information Categories and Metadata Quiz

Information Categories

• Structured information is organized and ordered, broken down into parts like facts or data. A sales transaction with fields for date, customer, item, and amount is an example. Relationships are straightforward, like a customer order relating to a customer record.
• Unstructured information has no inherent order or structure. Examples include assorted documents or items about a legal case (e.g., photos, handwritten notes, and articles). Relationships are difficult to identify.
• Semi-structured information has some structure but less than fully structured information. Web pages, with titles, subtitles, and content are examples. Relationships are easier to query and combine than unstructured data, but lack the strong structure of structured data.

Metadata

• Metadata describes data about data (clarifies information).
• Structured metadata for information describes the field definitions, tables, and relationships in information. Eg. a field called BirthDate is of data type Date/Time and format MM/DD/YYYY.
• Unstructured metadata describes properties of a document or resource. For a photo, metadata might include the date taken, photographer, keywords like beach or vacation.

Information Quality

• Accuracy: errors in dates, spelling, or prices affect quality (e.g., incorrect zip code).
• Precision: estimations with varying degrees of accuracy; "about 2 miles" for property surveys is inappropriate.
• Completeness: omission of certain information may need to be accounted for (e.g., house number to process an order)
• Consistency: conflicting definitions across different data gathering methods regarding information, resulting in incorrect or incomplete summaries.
• Timeliness: outdated data has less value than up-to-date data (e.g stock trading with a fraction of a second).
• Bias: lack of objectivity, may lead to misleading data like a manager trying to make a summary report seem greater, by including canceled orders
• Duplication: repetitive information, for example, a customer listed more than once in a database due to changes in address.

Early Database Management Approaches

• File processing systems: each department manages and stores its own records (e.g., payroll, HR). Problems arise from data redundancy, inconsistent definitions, and a lack of data integration.
• Hierarchical databases: tree-like structure for one-to-many relationships. Difficult to manage many-to-many relationships.
• Network databases: more flexible than hierarchical ones. They can handle many-to-many relationships more easily but are complex to manage.
• Relational databases: most widely used, organize data in tables linked by matching fields. Highly flexible and simple to manage complex relationships. More common in current systems.

Database Access Strategies

• Typical strategy: application interface (web-based forms) for users to enter, edit, or retrieve data securely.
• Other techniques: query languages (SQL), IVR systems, mobile apps, or natural language interfaces for specific tasks.

Database Administrator (DBA) Role

• Responsible for the efficient operation and management of databases. The key duties of DBAs include monitoring, optimizing, and troubleshooting systems, setting up new databases, and handling security issues. The responsibilities include resolving problems, building innovative applications and adhering to the company's database needs.
• Career outlook is strong, with projected growth and a median salary of $87,200.

Structured Query Language (SQL)

• Standard language for manipulating data in relational databases.
• Used for creating, reading, updating, and deleting information in databases.

Databases Without Boundaries

• Databases managed by parties outside the enterprise (e.g., Craigslist, Instagram, Google Person Finder).

Information Ownership Issues

• Conflicts arise from information ownership perspectives of different stakeholder groups,
• Top-level management requires high-level data insights.