Fraunhofer Chalmers Research Centre for Industrial Mathematics
Chalmers Publication LibraryNot a member yet
70439 research outputs found
Sort by
Evaluation of position and current sensor technologies for a PMSM used in automotive applications
No full text
Embracing the Himalayas - Investigating how a larger building complex can be positioned beautifully into an exceptional and challenging landscape
No full textModellering av SUC2-responsen i Snf1-signalvägen hos S. cerevisiae efter extern glukosminskning
No full textThe Snf1 pathway is an important part of the nutritional signaling network in yeast.
In this project the Snf1 pathway in S. cerevisiae was investigated using mathematical
modelling. Four different models of the expression of the invertase SUC2 after a shift
in the external glucose concentration from 4 % to 0.1 % were created. In addition to
Snf1 and SUC2 the modelled pathway also includes the transcription factor Mig1. The
models were made using ordinary differential equations.
In order to validate the models they were fitted to experimental data. The first two
models, model 0 and model 1, did not accurately describe the observed behavior of the
expression of SUC2 and were not analysed further. The other two models, however, did.
These were named the X-models, and are two simplifications of the same model, reduced
to become structurally identifiable. In the X-models Snf1 was black boxed and all the
unknown factors affecting the expression of SUC2 were lumped together in the state
variable X. One of the X-models, the lightly reduced version, captured the observed
behavior in the beginning of the experiment, whereas the other X-model, the heavily
reduced version, captured the stabilization of the expression of SUC2 which occured by
the end of the experiment.
Residual analysis was performed on the X-models and in order to investigate their ability
for prediction a numerical identifiability analysis was initiated. This, however, failed, as
the Hessian approximations could not be used to calculate the covariance matrices
Maskininlärning för klassificeringsalgoritmer
No full textThis report examines four basic machine learning algorithms for classification problems: linear regression,
the Precptron-algorithm, the Winnow-algorithm and a type of artificial neural network called
multi-layer Perceptron. The mathematical foundation of these algorithms are developed throughout the
report and provides the reader with a better understanding of the algorithms. Furthermore the algorithms
are implemented in MATLAB and the code is availible in appendix for reference.
The proportion of missclassified data is used as the main indicator of an algorithms performance,
however the precison and recall of the algorithms is also taken into account. Three different data sets are
tested: two data sets are simulated in MATLAB and a third dermatologic data set is gathered from a public
database. These data sets are used to evaluate the performance of the algorithms. The simulated sets are
chosen such that they introduce certain difficulties for the classification of the data.
Finally, two performance boosting algorithms are implemented and tested for each classification algorithm.
The result of using the boosting algorithms are compared to the results of not using them.
The median of the proportion of missclassified examples is for both simulated data sets 0 % for all
algorithms. Moreover the four algorithms achieves a median error of 2:78 % for the dermatologic data
set even if some spread and differences between the algorithms are noted
