Projekt

A machine learning approach to explain and predict melanosis in Atlantic salmon

Udbyder

Vejleder

Sted

København og omegn

Dark discolorations in salmon (dark spots) fillets are mainly caused by melanin deposits in the flesh. In the production of salmon the presences of these spots is very costly as the fillets become (partially) unusable. Through the past five years, producers of salmon have experienced a substantial increase in the fraction of salmons with dark spots in their production. Up to about 20% of the fish, have dark spots to some extent.

 

Knowledge of how and why these spots occur is scarce. Most of the existing studies look at one (or a few) potential causes at a time, in a controlled environment. Based on historical production data from Cermaq, one of the top three producers of salmon in the world, the student will take a machine leaning (ML) approach to broadening the knowledge of the nature of the dark spots. The student should consider existing hypothesis and how ML could support (or weaken) these. The project should focus on both prediction and interpretation. Finding associations in data that could be subject for future studies and/or could be incorporated in the workflows of Cermaq to reduce the problem.

 

Data from Cermaq database system (Fish Talk) throughout the past approx. 5 years is available. Data from other sources eg relevant weather data could be included.

 

It is the intention to publish the work. The student is encouraged to participate in writing a scientific article after finishing the master.

 

Supervisors:

Associate Professor Line H. Clemmensen, Statistics and Data Analysis Section, DTU Compute, lkhc@dtu.dk

PhD Student Kira Svendsen, Statistics and Data Analysis Section, DTU Compute, kdsv@dtu.dk

I samarbejde med

Cermaq Norway, Nofima

Søg i opslag
Kontakt

Virksomhed/organisation

DTU Compute

Navn

Kira Dynnes Svendsen

Stilling

Ph.d.-studerende

Mail

kdsv@dtu.dk

Vejleder-info

Kandidatuddannelsen i Informationsteknologi

Vejleder

Kira Dynnes Svendsen

ECTS-point

30 - 35

Type

Kandidatspeciale

Kandidatuddannelsen i Digitale Medieteknologier

Vejleder

Kira Dynnes Svendsen

ECTS-point

30 - 35

Type

Kandidatspeciale

Kandidatuddannelsen i Matematisk Modellering og Computing

Vejleder

Kira Dynnes Svendsen

ECTS-point

30 - 35

Type

Kandidatspeciale

OM DTU

DTU er et teknisk eliteuniversitet med international rækkevidde og standard. Vores mission er at udvikle og nyttiggøre naturvidenskab og teknisk videnskab til gavn for samfundet. 10.000 studerende uddanner sig her til fremtiden, og 5.700 medarbejdere har hver dag fokus på uddannelse, forskning, myndighedsrådgivning og innovation, som bidrager til øget vækst og velfærd.

Find os her

Anker Engelunds Vej 1
Bygning 101A
2800 Kgs. Lyngby


45 25 25 25

dtu@dtu.dk

CVR-nr. 30 06 09 46

Liste over EAN Numre

Job på DTU

Se alle jobs
 

loading..