Kapitel 9

Question 1

DBMS

Answer 1

Database management system handles the actual manipulation of the database.

Question 2

Association analysis

Answer 2

involves looking for links between data groups.

Question 3

Commit point

Answer 3

the point at which all the steps in a transaction have been recorded in the log. In the case of an equipment malfunction the dbms can use the information in its log to reconstruct the transactions that have been completed (commited) since the last backup was made.

Question 4

Schema

Answer 4

a description of the entire database structure that is used by the database software to maintain the database.

Question 5

Clustering

Answer 5

If a disproportionate number of keys happen to hash to the same bucket.

Question 6

Object-Oriented databases

Answer 6

This approach leads to a databse consisting of objects that are linked to each other to reflect their relationships. For example, an object oriented implementation of the eomployee dataset could consist of three classes (types of objects): Employee, job and assignment.

Question 7

Load factor

Answer 7

– a ratio that signifies well performing hash files, and the ratio of the number of records to the total record capacity of a file should remain below 50 percent.

Question 8

Data mining: Class discrimination

Answer 8

deals with identifying properties that divide two groups.

Question 9

Cascading rollback

Answer 9

when a rollback prompts several rollbacks due to the fact that different functions has been executed because of the first erroneous function.

Question 10

Application software

Answer 10

Handles the communication with the user of the database and may be quite complex, as exemplified by applications in which users access a database by means of a website. In that case the entire application layer consists of clients throughout the internet and a server that uses the database to fill the requests from the clients.

Question 11

Data mining: Cluster analysis

Answer 11

another form of data mining which seeks to discover classes. This differs from class description which seeks to discover properties of members within classes. Cluster analysis tries to fond properties of data items that lead to the discovery of groupings.

Question 12

Subschema

Answer 12

description of only that portion of the database pertinent to a particular user’s needs.

Question 13

Data mining

Answer 13

– differs from traditional databse interrogation in that data mining seeks to identify previously inknown patterns as opposed to traditional database inquiries that merely ask for the retrieval of stored facts. Data mining is practiced on static data collections called data warehouses. Two common forms of data mining are class description, class discrimination and cluster analysis

Question 14

Data mining: Class description

Answer 14

deals with identifying properties that characterize a given group of data items.

Question 15

SQL

Answer 15

Structured query language, a language that provide routines of combinations of the operations SELECT, PROJECT and JOIN. SQL forms the backbone of most relational database query systems.

Question 16

Outlier analysis

Answer 16

tries to identify data entries that do not comply with the norm.

Question 17

Database management systems

Answer 17

Consists of two major layers:Application layer and database management layer.

Question 18

Lossless or nonloss decomposition

Answer 18

When dividing a relation in a database into smaller relations does not caus loss of information.

Question 19

The relational database model

Answer 19

It portrays data as being stored in rectangular tables, called relations, which are similar to the format in which information is displayed by spreadsheet programs. A row in a relation is called a tuple. Columns in a relation are referred to as attributes because each entry in a column describes some characteristic or attribute of the entity represented by the corresponding tuple.

Question 20

Hashing, has system

Answer 20

Allows a record to be located by means of a key value, but rather than looking up the key in an index, hashing identifies the location of the record directly from the key.

The data storage space is divided into several sections called buckets, each of which is capable of holding several records. The records are dispersed among the buckets according to an algorithm tha tconverts key values into bucket number. Thes conversion from key values to bucket numbers is called hash function. Each record is stored in the bucket identified by this process. Therefore a record that has been placed in the structure can be retrieved by first appplying the hash function to the records identifying key to determine the appropriate bucket, then retrieving the contents of that bucket, and finallly searching through the data retrieved for the desired record.

Question 21

Indexed files

Answer 21

Sequential files are ideal for storing data that will be processed on the order in which the files entries are stored, however such files are inefficient when records within the file must be retrieved in an unpredictable order. A solution is to use an index for the file. An index contains a list of the keys stored in the file along with entries indicating where the record containing each key is stored. THus to find a particular record, one finds th eidentifying key in the index and then retrieves the block of information stored at the location associated with that key,

Question 22

Hash file and hash table

Answer 22

When hashing is applied to a storage structure in mass storage, the result is called a hash file. When applied to a storage structure within main memory the result is usually called a hash table.

Question 23

Sequential pattern analysis

Answer 23

Sequential pattern analysis tries to identify patterns of behavious over time.