Semantische Suche
Freitag, 2. August 2019, 11:30 Uhr
iCal (Download)
Ort: Raum 348 (Gebäude 50.34)
Webkonferenz: {{{Webkonferenzraum}}}
(Keine Vorträge)
Freitag, 9. August 2019, 11:30 Uhr
iCal (Download)
Ort: Raum 010 (Gebäude 50.34)
Webkonferenz: {{{Webkonferenzraum}}}
| Vortragende(r) | Andreas Foitzik |
|---|---|
| Titel | Enhancing Non-Invasive Human Activity Recognition by Fusioning Electrical Load and Vibrational Measurements |
| Vortragstyp | Masterarbeit |
| Betreuer(in) | Klemens Böhm |
| Vortragsmodus | |
| Kurzfassung | Professional installation of stationary sensors burdens the adoption of Activity Recognition Systems in households. This can be circumvented by utilizing sensors that are cheap, easy to set up and adaptable to a variety of homes. Since 72% of European consumers will have Smart Meters by 2020, it provides an omnipresent basis for Activity Recognition.
This thesis investigates, how a Smart Meter’s limited recognition of appliance involving activities can be extended by Vibration Sensors. We provide an experimental setup to aggregate a dedicated dataset with a sampling frequency of 25,600 Hz. We evaluate the impact of combining a Smart Meter and Vibration Sensors on a system’s accuracy, by means of four developed Activity Recognition Systems. This results in the quantification of the impact. We found out that through combining these sensors, the accuracy of an Activity Recognition System rather strives towards the highest accuracy of a single underlying sensor, than jointly surpassing it. |
Freitag, 16. August 2019, 11:30 Uhr
iCal (Download)
Ort: Raum 348 (Gebäude 50.34)
Webkonferenz: {{{Webkonferenzraum}}}
| Vortragende(r) | Huijie Wang |
|---|---|
| Titel | Predictability of Classfication Performance Measures with Meta-Learning |
| Vortragstyp | Bachelorarbeit |
| Betreuer(in) | Jakob Bach |
| Vortragsmodus | |
| Kurzfassung | Choosing a suitable classifier for a given dataset is an important part in the process of solving a classification problem. Meta-learning, which learns about the learning algorithms themselves, can predict the performance of a classifier without training it. The effect of different types of performance measures remains unclear, as it is hard to draw a comparison between results of existing works, which are based on different meta-datasets as well as meta-models. In this thesis, we study the predictability of different classification performance measures with meta-learning, also we compare the performances of meta-learning using different meta-regression models. We conduct experiments with meta-datasets from previous studies considering 11 meta-targets and 6 meta-models. Additionally, we study the relation between different groups of meta-features and the performance of meta-learning. Results of our experiments show that meta-targets have similar predictability and the choice of meta-model has a big impact on the performance of meta-learning. |
Freitag, 23. August 2019, 11:30 Uhr
iCal (Download)
Ort: Raum 348 (Gebäude 50.34)
Webkonferenz: {{{Webkonferenzraum}}}
(Keine Vorträge)
Freitag, 30. August 2019, 11:30 Uhr
iCal (Download)
Ort: Raum 348 (Gebäude 50.34)
Webkonferenz: {{{Webkonferenzraum}}}
| Vortragende(r) | Albu Dumitru-Cristian |
|---|---|
| Titel | Implementation and Evaluation of CHQL Operators in Relational Database Systems to Query Large Temporal Text Corpora |
| Vortragstyp | Bachelorarbeit |
| Betreuer(in) | Jens Willkomm |
| Vortragsmodus | |
| Kurzfassung | Relational database management systems have an important place in the informational revolution. Their release on the market facilitates the storing and analysis of data. In the last years, with the release of large temporal text corpora, it was proven that domain experts in conceptual history could also benefit from the performance of relational databases. Since the relational algebra behind them lacks special functionality for this case, the Conceptual History Query Language (CHQL) was developed. The first result of this thesis is an original implementation of the CHQL operators in a relational database, which is written in both SQL and its procedural extension. Secondly, we improved substantially the performance with the trigram indexes. Lastly, the query plan analysis reveals the problem behind the query optimizers choice of inefficient plans, that is the inability of predicting correctly the results from a stored function. |