0x010088F6312065A70A44B630A6848498EAD80093CFEAFE161DC94D83C7C5C9A5A748B8

For the first time since the Covid pandemic, all lectures and presentations could be held onsite, with the slides published afterwards on Blackboard together with course information, some literature, etc. Blackboard was also utilised for continuous feedback on lectures and exercises. A new course book was used this year, Getting Started with Natural Language Processing by Ekaterina Kochmar (Manning Publishing, September 2022), complemented by the hands-on Natural Language Processing with Python by Steven Bird, Ewan Klein &amp; Edward Loper (3 ed., available for free online). The final exam was onsite, with the written exam accounting for 25% of the course grade and the individual student projects (including a written report, oral presentation, and code) accounting for 75%. As a result of student feedback from last year, this year’s projects were allowed to be carried out either by students working alone or two students working together. In addition, there were four obligatory coding exercises, mainly introducing the techniques used in the field and laying the foundations for the project work. Course StatisticsIn total 204 students were officially enrolled over time this year (156 in 2022, 146 in 2021, 117 in 2020, 48 in 2019; and 10 in 2018, the first year the course was given). At the end of the course, 78 were enrolled (89 in 2022, 83 in 2021, 81 in 2020, 37 in 2019), but with only 46 actually participating in the lab exercises (69 in both 2022 and 2021, 53 in 2020, 22 in 2019). Of those, 5 withdrew during the course (8 in 2022, 11 in 2021, 10 in 2020, 3 in 2019). A further 8 did not submit the project assignment and one student was expelled, so 32 students completed the course (45 in 2022, 49 in 2021, 33 in 2020, 18 in 2019, 3 in 2018). The average grade points among those completing the course was 73.44/100 (79.33 in 2022, 78.93 in 2021, 83.12 in 2020, 76.28 in 2019, 86.00 in 2018).

In spite of the staff vigorously trying to recruit volunteers for a reference group, no student was willing to participate it, so the course was mainly evaluated through an online questionnaire, which was sent out to all 204 students who at one time or another had enlisted for the course. Hence students who only registered but never participated were also invited, as well as those who dropped out during the course, and those who completed it. The idea with inviting all registered students was to potentially get feedback as to some students decided to leave the course or prioritise other courses, in addition to getting feedback from the students who actually followed the course to the end. Of the replying students that had not participated at all in the course, two gave as reasons “dropped out due to an overlap” and “We were told from NTNU that students at the master program in business administration with specialization in Business Analytics could take this course, but it was unfortunately very clear from the information on blackboard that we were not welcome.” A student who dropped out after a few exercises gave as reason “Prioritised other courses. The course seemed too difficult / too much work”. The students who completed the course and took the questionnaire were asked how well they agreed with 25 statements regarding the course range 1-5. 1=totally disagree; 5=totally agree), as well as three free-text questions (“What do you think was the best thing about this course”, “What do you think is most in need of improvement?” and “Any other comments or suggestions?”). The students were overall satisfied with the course assessment and workload, and stated that the course had helped improve their problem solving skills. The explanation of course requirements, the exercises and lectures all got above-average ratings, but the course literature got mixed feedback. The free text feedback from the students on the questionnaire:What do you think was the best thing about this course?·      I really liked the project with presentasjon and «defending».·      I was a bit dissapointed how little the opponent had to do in the end, and would have enjoyed a more in depth discussion between opponent/presenter Any other comments or suggestions?·      As far as the question about memorization I felt that that was the case for the final exam, but not for the project. ·      I feel like the majority of what i Learned in the course was in the project.

NTNU:TIA:REST

Institutt for datateknologi og informatikk

Department of Computer Science

Ingen trykte eller håndskrevne hjelpemidler tillatt. Bestemt, enkel kalkulator tillatt.

Engelsk

Norsk

Mer om vurdering

Skriftlig eksamen (25 %) og arbeider (75 %) gir grunnlag for sluttkarakter i emnet. Sensur angis med bokstavkarakterer både på de enkelte delene og på hele faget. Ved utsatt eksamen (kontinuasjonseksamen) kan skriftlig eksamen bli endret til muntlig eksamen. Øvinger fra tidligere kan godkjennes etter avtale med faglærer.

Anbefalte forkunnskaper

Emnet er åpent for alle studenter, men forutsetter kunnskap i Pythonprogrammering minst på et nivå tilsvarende emnet TDT4110 ("Informasjonsteknologi, grunnkurs"); programmeringskunnskap på nivå som en datateknikk-/informatikkstudent i 3. klasse anbefales. Grunnleggende kunnskap om maskinlæring er ikke nødvendig, men anbefales. Det forutsettes ingen kunnskap om lingvistikk eller om språkteknologi.

Forkunnskapskrav

Kunnskap i Pythonprogrammering minst på et nivå tilsvarende TDT4110 ("Informasjonsteknologi, grunnkurs").

Faglig innhold

Emnet omfatter: menneskelige språk og språkbehandling, maskinlæring og statistiske metoder for språkførståelse og tekstanalyse, tekstklassifisering og dyp læring, sentimentanalyse og meninggruvedrift, informasjonaksess og tekstgruvedrift, konverserende agenter, maskinoversettelse, semantikk for naturlige språk, datamaskinell lingvistisk kreativitet.

Kursmateriell

Artikler, støtteark og bøker på nettet. Mer detaljert informasjon vil bli gitt ved semesterstart.

Læringsutbytte

Studentene skal ha en grunnleggende og praktisk forståelse av aktuelle rammeverk og metoder for tekstanalyse og naturlig språk-forståelse, spesiell bruk av maskinlæringsmetoder for tekstanalyse.

Læringsformer og aktiviteter

Forelesninger og øvinger. Emnet krever en godkjent prosjektrapport med teoretisk og eksperimentelt innhold. Forelesningene vil bli gitt på engelsk, hvis ikke alle studenter snakker norsk.

Øvinger

Intelligent tekstanalyse og språkforståelse

Datateknikk og informasjonsvitenskap

Anvendt informasjons- og kommunikasjonsteknologi

Informatikk

Kunnskapssystemer

Informasjonssystemer

Datateknikk

Anvendt lingvistikk

Anvendt språkvitenskap

Informasjonsteknologi og informatikk

Kommunikasjon og informasjonsvitenskap

Språk, logikk og informasjonteknologi

Master of Science in Informatics

Informatikk - bachelorstudium

Datateknologi - masterstudium (2-årig)

Datateknologi - masterstudium (5-årig)

Web/Lists(guid'8b698a2d-6ad0-4f7d-b3be-219cc1979f02')/Items(23498)/FieldValuesAsText

Mappevurdering gir grunnlag for sluttkarakter i emnet. I mappen inngår avsluttende skriftlig eksamen (25%) og arbeider (75%). Resultatet for delene angis i %-poeng, mens sensur for hele mappen (sluttkarakteren) angis med bokstavkarakter. Ved utsatt eksamen (kontinuasjonseksamen) kan skriftlig eksamen bli endret til muntlig eksamen. Dersom studenten også etter utsatt eksamen har sluttkarakteren F/ikke-bestått, må studenten gjenta hele emnet neste studieår. Arbeider som teller med i sluttkarakteren må gjentas.

Emnet er åpent for alle studenter, men forutsetter kunnskap i Pythonprogrammering minst på et nivå tilsvarende emnet TDT4110 ("Informasjonsteknologi, grunnkurs"); programmeringskunnskap på nivå som en datateknikk-/informatikkstudent i 3. klasse anbefales. Grunnleggende kunnskap om maskinlæring er ikke nødvendig, men anbefales. Det forutsettes ingen kunnskap om lingvistikk eller om språkteknologi.

TDT4110 ("Informasjonsteknologi, grunnkurs") eller tilsvarende.

* "Applied Text Analysis with Python" av Benjamin Bengfort, Tony Ojeda og Rebecca Bilbro (O'Reilly Media, juni, 2018). * "Natural Language Toolkit" (http://www.nltk.org/). Mer detaljert informasjon vil bli gitt ved semesterstart.

Web/Lists(guid'8b698a2d-6ad0-4f7d-b3be-219cc1979f02')/Items(17986)/FieldValuesAsText

Intelligent tekstanalyse og språkforståelse - TDT4310