1.0 - 8.00 - NERExtractor Example - Teradata Vantage

Teradata® Vantage Machine Learning Engine Analytic Function Reference

Product
Teradata Vantage
Release Number
1.0
8.00
Release Date
May 2019
Content Type
Programming Reference
Publication ID
B700-4003-098K
Language
English (United States)

Input

  • Input table: ner_sports_test2, which contains text to analyze.
  • Rules table: rule_table, which is preinstalled on the ML Engine.
  • Model file: ner_model.bin, output by NERTrainer Example.
ner_sports_test2
id content
528 email sports@espn.com to contact for all sport info
529 email cricket@espn.com to contact for all cricket info
530 email tennis@espn.com to contact for all tennis info
531 1= Igor Trandenkov (Russia) 5.86
532 3. Maksim Tarasov (Russia) 5.86
533 4. Tim Lobinger (Germany) 5.80
534 5. Igor Potapovich (Kazakstan) 5.80
535 6. Jean Galfione (France) 5.65
536 7. Pyotr Bochkary (Russia) 5.65
537 8. Dmitri Markov (Belarus) 5.65
583  GENEVA 1996-08-30
584  UEFA came down heavily on Belgian club Standard Liege on Friday for disgraceful behaviour in an Intertoto final match against Karlsruhe of Germany .
585 The Belgian club were fined 25
586 He was sent off for insulting the referee and then urged his team mates to protest .
587  Roberto Bisconti will be sidelined for six Euro ties after pushing the referee in the back as he protested about a Karlsruhe goal
588  Karlsruhe won the August 20 match 3-1 thanks to two late goals .
589 They took the tie 3-2 on aggregate and qualified for the UEFA Cup .
591 ATHLETICS - HARRISON
592  MONTE CARLO 1996-08-30
593 Olympic champion Kenny Harrison and world record holder Jonathan Edwards will both take part in a triple jump competition at the Solidarity Meeting for Sarajevo on September 9 .
594 The International Amateur Athletic Federation said on Friday that a schedule reshuffle had allowed organisers to hold a men s triple jump as well as the women s long jump on the one usable runway at the war-devastated Kosevo stadium .
595 Atlanta Games silver medal winner Edwards has called on other leading athletes to take part in the Sarajevo meeting -- a goodwill gesture towards Bosnia as it recovers from the war in the Balkans -- two days after the grand prix final in Milan .
596  Edwards was quoted as saying : What type of character do we show by going to the IAAF Grand Prix Final in Milan where there is a lot of money to make but refusing to make the trip to Sarajevo as a humanitarian gesture ?
598 SOCCER - BARATELLI TO COACH NICE .
599  NICE
600 Former international goalkeeper Dominique Baratelli is to coach struggling French first division side Nice
601  Baratelli
602  Nice have been unable to win any of their four league matches played this season and are lying a lowly 18th in the table .
rule_table
type regex
email [\w\-]([\.\w])+[\w]+@([\w\-]+\.)+[a-zA-Z]{2,4}

SQL Call

SELECT * FROM NERExtractor (
  ON ner_sports_test2 PARTITION BY ANY
  ON rule_table AS rules DIMENSION
  USING
  TextColumn ('content')
  ModelFiles ('ner_model.bin')
  ShowContext (2)
  Accumulate ('id')
) AS dt ORDER BY id, sn;

Output

id sn entity type start end context approach
528 1 sports@espn.com email 2 2 ... email sports@espn.com to contact RULE
529 1 cricket@espn.com email 2 2 ... email cricket@espn.com to contac RULE
530 1 tennis@espn.com email 2 2 ... email tennis@espn.com to contact RULE
531 1 Igor Trandenkov PER 2 3 ... 1= Igor Trandenkov (Russia) 5.86 CRF
532 1 Maksim Tarasov PER 2 3 ... 3. Maksim Tarasov (Russia) 5.86 CRF
533 1 Tim Lobinger PER 2 3 ... 4. Tim Lobinger (Germany) 5.80 CRF
534 1 Igor Potapovich PER 2 3 ... 5. Igor Potapovich (Kazakstan) 5.80 CRF
535 1 Jean Galfione PER 2 3 ... 6. Jean Galfione (France) 5.65 CRF
536 1 Pyotr Bochkary PER 2 3 ... 7. Pyotr Bochkary (Russia) 5.65 CRF
537 1 Dmitri Markov PER 2 3 ... 8. Dmitri Markov (Belarus) 5.65 CRF
583 1 GENEVA LOC 1 1 ... ... GENEVA 1996-08-30 ... CRF
584 1 Standard Lieg PER 8 9 Belgian club Standard Liege on Friday CRF
587 1 Roberto Bisconti PER 1 2 ... ... Roberto Bisconti will be CRF
591 1 HARRISON PER 3 3 ATHLETICS - HARRISON ... ... CRF
592 1 MONTE CARLO PER 1 2 ... ... MONTE CARLO 1996-08-30 ... CRF
593 1 Kenny Harrison PER 3 4 Olympic champion Kenny Harrison and world CRF
593 2 Jonathan Edwards PER 9 10 record holder Jonathan Edwards will both CRF
596 1 What ORG 7 7 saying : What type of CRF
598 1 BARATELLI TO PER 3 4 SOCCER - BARATELLI TO COACH NICE CRF
599 1 NICE PER 1 1 ... ... NICE ... ... CRF
600 1 Dominique Baratelli PER 4 5 international goalkeeper Dominique Baratelli is to CRF
600 2 Nice PER 14 14 division side Nice ... ... CRF
601 1 Baratelli PER 1 1 ... ... Baratelli ... ... CRF