IBDP Computer Science A3.4 Alternative databases and data warehouses HL Paper 2 - New Syllabus

Question 

A characteristic of an object-oriented database is object hierarchy. This is where objects are organized into classes and subclasses.

(a)
(i) Identify three other characteristics of an object-oriented database. [3]

A multi-dimensional database could offer significant advantages for analysing the dental treatment data of patients.

(ii) Describe one advantage of having dental treatment data organized into dimensions. [3]

The dental clinic uses data matching to review patients’ past medical treatments.

(b) Explain one reason why data matching is an appropriate method of reviewing patients’ past medical treatments. [3]

The dental clinic has decided to use forecasting, a sub-branch of data mining, in its database.

(c) Outline two research areas to which the dental clinic could apply forecasting. [4]

The dental clinic has decided to apply database segmentation to focus on specific patient groups.

(d) Justify applying database segmentation to focus on specific patient groups. [4]

The dental regulatory authority uses a data warehouse to store the treatment data from multiple clinics.

(e) Explain why the three processes of extract, transform, and load (ETL) need to be applied to treatment data. [3]

Most-appropriate topic code

A3.4: Alternative databases and data warehouses [HL only] — parts (a) and (e)
A4.1: Machine learning fundamentals — parts (b) and (c)
A3.4: Alternative databases and data warehouses [HL only] — part (d)
▶️ Answer/Explanation

(a) (i)
For the correct answer:

  • Unique object identity: each object has a unique identifier.
  • Encapsulation: data and behaviour, including methods, are contained within an object.
  • Complex objects and data structures: an object-oriented database can represent structures such as trees, graphs and networks.

Other valid characteristics include polymorphism, dynamic binding, object persistence and object-level querying.

(a) (ii)
For the correct answer:

Organizing dental treatment data into dimensions allows the data to be analysed from different perspectives, such as demographics, location, date and time, or treatment type. This can reveal useful patterns and trends that support better patient care.

For example, analysing treatment data by age and treatment type could allow the clinic to identify common dental problems in particular patient groups and tailor treatments accordingly.

(b)
For the correct answer:

Data matching can combine and compare treatment information from different sources, such as patient records, treatment logs and pharmacy records. This provides a more complete view of a patient’s medical history, which can assist with future diagnosis and treatment.

It can also improve patient safety. For example, matching historical medical records could identify an allergy or previous adverse reaction, helping the dentist avoid an inappropriate treatment.

(c)
For the correct answer:

Dental services demand forecasting: historical data can be used to predict future patient demand for particular dental services. This can help the clinic plan resources and staffing.

Epidemiological trends: forecasting can predict trends in dental diseases or conditions. This can assist with planning preventive care and public-health responses.

Other valid research areas include financial forecasting, patient appointment/no-show analysis and staff scheduling.

(d)
For the correct answer:

  • Segmentation allows patients to be grouped according to characteristics such as age, treatment type or geographic location.
  • The clinic can identify common dental problems that are particularly prevalent within each group.
  • These findings can be used to develop targeted preventive measures and tailor treatments to the needs of specific patient groups.
  • This can lead to more personalized dental care and more efficient use of resources because different groups may require different services.

(e)
For the correct answer:

Extract: Data must first be collected from the different clinics and their sources, such as databases, cloud storage or spreadsheets. Different extraction processes may be needed for different sources and file formats.

Transform: The extracted data must be converted into a consistent format suitable for the data warehouse. The data can also be cleaned by correcting inaccuracies, removing duplicates and dealing with missing values to maintain data integrity.

Load: The transformed data must then be transferred into the appropriate data warehouse schema, such as its tables, dimensions and facts. The loading process should be checked to ensure that all expected data has been successfully written to the warehouse.

Scroll to Top