INFORMATİON TECHNOLOGİES

Data can be referred to as raw, unorganized facts and concepts which need to be processed. Information is organized, categorized, contextualized data. The correct option is C.

Managing data is important and provides some advantages described as follows:

The correct option is B.

Some of the topics in data management are:

The correct option is B.

The conceptual design provides a conceptual schema of DB. It allows representing entities and relationships between them in high-level view models. Entity-Relationship model is one of the popular data model used for conceptual design. It represents data as entities, attributes, and relationships. The correct option is D.

The steps of KDD is as follows:

The correct option is D.

Knowledge discovery is referred as a process of extracting interesting and useful information from huge amount of data which can be from structured and unstructured sources. Knowledge discovery from data (KDD) draws the interests of many researchers studying on machine learning, pattern recognition, data mining, etc. The correct option is A.

Data mining is the main step in KDD process and discovers data patterns from various data repositories. Data mining functionalities specify the type of data for data mining tasks. The correct option is C.

Bayesian classifier is one of the most known classification techniques for predictive data mining tasks. The others are used for descriptive data mining tasks. The correct option is B.

Mining clusters bring together similar objects which do not have a class label. The correct option is A.

Metadata is generally referred as data about other data or catalog for a data. It describes, identifies, explains, manages, and uses a data resource. Metadata can be grouped as descriptive, structural and administrative metadata. Descriptive metadata identifies a data resource. Administrative metadata defines the ways of managing a data resource. Structural metadata describes relationships between parts of a composite object. The correct option is E.

Data needs to be defined, its structure must be described. Data is stored to databases for online transactional purposes, creating, modifying, deleting, retrieving data. Data warehousing keeps historical data for preprocessing, mining, analyzing, reporting data. Data security topics such as accessing to data, privacy of data, erasing data must be managed. Data quality is important for decision-making. Data quality must be managed since data quality affects quality of extracted knowledge. Data architecture is the set of rules, models, methodologies to determine data flow for enterprises’ applications, to manage data in rest for enterprises’ information technology systems and so on. Data architecture must be managed to design, deploy and modify it. Therefore, the correct option is D.

An entity is a real-world object, whereas an attribute represents a property describing the related entity. An attribute can be any type of composite, derived, simple and multivalued. A simple attribute is the attribute which is not divisible. To illustrate, an id property for a book is simple attribute. A composite attribute symbolizes the collection of simple attributes. Thus, it can be divided into sub-attributes. To illustrate, name attribute is a composite attribute which is composed of simple attributes such as first name, middle name and last name. A multi-valued attribute contains more than one entry for a specific entity. To illustrate, birthdate of an author is single-valued because there is only one birthdate for the corresponding author whereas, phone number is a multi-valued attribute if there are more than one valid phone number for the related author. An attribute is derived attribute if it can be derived from other described attributes. To exemplify, age is a derived attribute for an author because it can be derived from birthdate attribute. Therefore, the correct option is E.

The number of attendant entities of a relationship represents the degree of the corresponding relationship. If there are two participated entities in a relationship type, that relationship is referred as binary relationship. A ternary relationship has three participants whereas n-ary relationship has n participants. For a binary relationship, cardinality ratio can be one of the following options: one-to-one (1:1), one-to-many (1:N), many-to-one (N:1) and many-to-many (N:M). Two types of participation constraints exist. If every entity in an entity set is participated in the relationship type then the participation for corresponding entity type is total. If there are some entities of an entity type which do not participate in the relationship type, then that entity type is partially participated. Therefore, the correct option is E.

KDD is a collection of seven steps:

Therefore, the correct option is A.

Discretization is the work of dividing the range of values of a continuous attribute into intervals. Data cleaning includes processes of filling missing values, correcting inconsistencies, identifying outliers and getting rid of them. Huge amount of data makes mining data in coherent time impossible. Thus data reduction techniques are utilized. Reducing number of tuples, dimensionality reduction and etc. are utilized techniques for data reduction. Data integration merges data from different sources such as database, data flats etc. into a compatible data store. Data transformation includes the technique of aggregation, generalization, normalization and attribute construction. Therefore, the correct option is D.

Data classification is a two-step process as training and prediction. In training step, a classifier model is created by analyzing data objects and their class labels. In a prediction step, unknown class labels of data objects are predicted using the generated classifier model. Decision tree induction is one of the mostly known classification techniques. Thus, it falls under the category of predictive data mining tasks. In training step, it generates a decision tree based on class labels of trained data objects. In a decision tree, each node represents a test on an attribute of data, each edge of that node represents a value for that attribute and each leaf represents a class label. Generating a decision tree is generally as a two-step process as tree construction and tree pruning. Therefore, the correct option is E.

If-Then rule, k-nearest neighbor, Bayesian classifiers and Naïve Bayesian classifiers are used for classification purposes. On the other hand, regression is one of the mostly known prediction algorithms. Therefore, the correct option is A.

Mining of Frequent Patterns is the process of mining patterns that frequently occurs in data. Data can be described with classes or concepts. These descriptions are inferred with data characterization and data discrimination. Data characterization is the process of summarizing the general properties of a class whereas data discrimination is the process of comparing class properties with features of other one or more classes. Association is referred as patterns frequently purchased together. Mining association is the process of discovering relations among data. Correlations represent relations between items or their attributes. Two items can be positively or negatively correlated. Moreover, there may be no correlation between two items as a result of mining correlation. Clustering brings together similar objects which do not have a class label. Therefore, the correct option is E.

Partitioning-based methods divide n objects into k partitions where k is smaller than or equal to n. In partitioning-based techniques, each partition contains at least one object and each object belongs to only one partition. Objects are divided hierarchically in hierarchical- based clustering. Hierarchical-based clustering methods can be classified as agglomerative and divisive approaches. Initially, each object represents a cluster in agglomerative approaches, then close clusters are merged consecutively until a termination condition holds or all objects or clusters are collected into one cluster. In divisive approaches, all objects are grouped into one cluster. Then, a cluster is divided into smaller clusters in each consecutive step until each object is assigned to one cluster or a termination condition holds. Density-based clustering methods group objects into clusters based on densities of clusters to represent clusters with arbitrary shape. An object in a cluster must have at least minimum number of neighbors in the neighborhood of a given radius. In model-based methods, objects are assigned to clusters based on defined models of clusters. Therefore, the correct option is C.

Metadata is generally referred as data about other data or catalog for a data. It describes, identifies, explains, manages, and uses a data resource. Metadata can be grouped as descriptive, structural and administrative metadata. Descriptive metadata identifies a data resource. Administrative metadata defines the ways of managing a data resource. Structural metadata describes relationships between parts of a composite object. Metadata can generally be stored in various storage structures such as a field of relational database’s table, XML files, RDF models. Therefore, the correct option is D.