Options

Inconsitency and Incompleteness

Juju147Juju147 Member Posts: 5 Contributor II
edited November 2018 in Help
Hi everyone,

I am new user of Rapidminer and I like to konw if there is some operators or some process made to show inconsistency or incompleteness in a database. ???

Sincerly,

JuJu147

Answers

  • Options
    MariusHelfMariusHelf RapidMiner Certified Expert, Member Posts: 1,869 Unicorn
    Hey Juju,

    how do you define inconsistency and incompleteness?

    Best regards,
    Marius
  • Options
    Juju147Juju147 Member Posts: 5 Contributor II
    Hi Marius,

    For me inconsistency is some mistakes in database wich can be bad for the integrity. For example, orthograph mistakes in the name of people. The architecture of my database is the following :
    /phdthesis/#id |  /phdthesis/@key| /phdthesis/@mdate| /phdthesis/author| /phdthesis/ee| /phdthesis/isbn| /phdthesis/school | /phdthesis/title | /phdthesis/url | /phdthesis/year


    and I am trying to underline the fact that wrong orthograph in the name of the autor lead inconsistency problem.

    Thank you for your time,

    Sincerly,

    Ju
  • Options
    MariusHelfMariusHelf RapidMiner Certified Expert, Member Posts: 1,869 Unicorn
    Hi Ju,

    unfortunately finding spelling errors is not the main focus of RapidMiner. It would be possible to create a process to do something like that, but unless you want to do further data processing, ETL and data analysis I would probably use another tool for finding the orthographic mistakes.

    Best regards,
    Marius
  • Options
    Juju147Juju147 Member Posts: 5 Contributor II
    Thanks for your answer

    I have a question the k-NN operator... How does it work ? because i think i can make a comparaison between to close cell with that one right ?

    Ju
  • Options
    MariusHelfMariusHelf RapidMiner Certified Expert, Member Posts: 1,869 Unicorn
    Hm, no, not really. The k-NN operator is used for classification problems. You are right in that it creates its predictions by comparing new instances to the nearest, i.e. most similar other instances, but the user does not have access to these comparisons, but only sees the final prediction.

    In any case, unfortunately, it cannot be used to find spelling errors.

    Best regards,
    Marius
Sign In or Register to comment.