Post by account_disabled on Mar 5, 2024 1:12:18 GMT -6
Data migration The stages to guarantee the integrity, reliability and quality of the data in a migration process are the three that we point out below: I. Research The priority of any data migration project must be to guarantee the quality of the information , since business intelligence is based on operating on reliable and complete data . In the data migration project, it is essential, as always, to obtain true and accurate information. It is important, therefore, to investigate all the data that will be subject to the migration process, not to discard or use greater resources for the investigation of some data relegating others, to have all possible inputs (data owners, users who interact with it...), contrast the information obtained and validate the conclusions reached. At this stage you must be able to discover potential anomalies in the data , achieve 100% visibility of the free content fields, identify default and invalid values, reveal undocumented business rules , guarantee the veracity of the data contained in the fields.
Fields that will be used for the matching criteria, and understand the data in context. II. Standardization It means delving into the knowledge of the data from a logical perspective and minimizing its abstract part to facilitate its transfer under control conditions. It is motivated by different needs that arise throughout the data migration process. Standardization implies optimal conditions in terms of the incorporation of a highly flexible pattern Chile Mobile Number List recognition language , the use of specific rules for names and surnames, addresses or dates, the division of data by their nature (names, type of road, street name, street number...), the normalization of data writing and the parameterization of classification and standardization tables . III. Pairing Also known as Data Matching , it consists of making a comparison of the data that will be the subject of migration with other data collected in a knowledge database. It is previously necessary to define an acceptance percentage that is considered valid to establish a pairing policy that sets the guidelines. Data pairing is necessary, in addition to a good data migration, to trust its consistency and integrity once the migration is complete.
The purpose of data matching techniques is to identify possibly matching records, establish relationships between records in different files, document in deterministic decision tables that must result in a match (field comparison, letter grade assigned, letter grade combination, letter grade assignment to file...) and include the probabilistic matching of records, which are resolved in the statistical probability of a match (they are measured with the evaluation of the fields by the degree of agreement, the assignment of weights that represents the content by value, and the sum of weights to assign a total weight). For matching to be fully effective, it must be complemented with data cleaning by defining standards that dictate which data is correct and which is not. Cloud-based deployment option: which provides all the functionality of the on-premise solution with a cloud-based deployment option and: It does not affect current or future capacity, options or integrity of the work environment. Simplifies data exchange. Allows you to choose between centralized or distributed planning deployments. TM1 can also work in offline mode. Related posts: BI trends: what is TM1 TM1 Cognos Data manager: features and benefits TM1 reporting: practical tips.
Fields that will be used for the matching criteria, and understand the data in context. II. Standardization It means delving into the knowledge of the data from a logical perspective and minimizing its abstract part to facilitate its transfer under control conditions. It is motivated by different needs that arise throughout the data migration process. Standardization implies optimal conditions in terms of the incorporation of a highly flexible pattern Chile Mobile Number List recognition language , the use of specific rules for names and surnames, addresses or dates, the division of data by their nature (names, type of road, street name, street number...), the normalization of data writing and the parameterization of classification and standardization tables . III. Pairing Also known as Data Matching , it consists of making a comparison of the data that will be the subject of migration with other data collected in a knowledge database. It is previously necessary to define an acceptance percentage that is considered valid to establish a pairing policy that sets the guidelines. Data pairing is necessary, in addition to a good data migration, to trust its consistency and integrity once the migration is complete.
The purpose of data matching techniques is to identify possibly matching records, establish relationships between records in different files, document in deterministic decision tables that must result in a match (field comparison, letter grade assigned, letter grade combination, letter grade assignment to file...) and include the probabilistic matching of records, which are resolved in the statistical probability of a match (they are measured with the evaluation of the fields by the degree of agreement, the assignment of weights that represents the content by value, and the sum of weights to assign a total weight). For matching to be fully effective, it must be complemented with data cleaning by defining standards that dictate which data is correct and which is not. Cloud-based deployment option: which provides all the functionality of the on-premise solution with a cloud-based deployment option and: It does not affect current or future capacity, options or integrity of the work environment. Simplifies data exchange. Allows you to choose between centralized or distributed planning deployments. TM1 can also work in offline mode. Related posts: BI trends: what is TM1 TM1 Cognos Data manager: features and benefits TM1 reporting: practical tips.