|
|
Document embedding using Transformers
Burian, David ; Libovický, Jindřich (advisor) ; Variš, Dušan (referee)
We develop a method to train a document embedding model with an unlabeled dataset and low computational resources. Using teacher-student training, we distill SBERT's capacity to capture text structure and Paragraph Vector's ability to encode extended context into the resulting embedding model. We test our method on Longformer, a Transformer model with sparse attention that can process up to 4096 tokens. We explore several loss functions for the distillation of knowledge from the two teachers (SBERT and Paragraph Vector) to our student model (Longformer). Throughout experimentation, we show that despite SBERT's short maximum context, its distillation is more critical to the student's performance. However, the student model can benefit from both teachers. Our method improves Longformer's performance on eight downstream tasks, including citation prediction, plagiarism detection, and similarity search. Our method shows excep- tional performance with few finetuning data available, where the trained student model outperforms both teacher models. By showing consistent performance of differently con- figured student models, we demonstrate our method's robustness to various changes and suggest areas for future work. 1
|
|
|
Marketing strategy proposal
Menšík, Jakub ; Burian, David (referee) ; Chalupský, Vladimír (advisor)
On grounds of a performed research the autor hereof offers a suitable marketing strategy which would lead to more efficient sales of goods produced by company PLEAS a.s. Herein, the theoretical knowledge is applied to a real situaution of the marketing research and the proposals resulting thereof are then analyzed and presented.
|
|
|
Animated visualizations for IVIS framework
Burian, David ; Bureš, Tomáš (advisor) ; Kruliš, Martin (referee)
With the growing volumes of collected data, their presentation and understanding have become much more critical. Visualization frameworks, such as IVIS, continuously develop new ways of graphically presenting data in order to, among other things, improve user experience and data insight. In this thesis, we develop an animation extension for IVIS, which creates a new, interactive and engaging way of exploring data. The resulting animation can play back either time- series or real-time data visualizing their evolution over a period of time. The animation extension relies on already existing IVIS's visualization components. We also focus on developing a new interface allowing the users to control the animation. As part of this thesis, we have designed sample dashboards showcasing the possibilities of the implemented solution. 1
|
|
|
Marketing strategy proposal
Menšík, Jakub ; Burian, David (referee) ; Chalupský, Vladimír (advisor)
On grounds of a performed research the autor hereof offers a suitable marketing strategy which would lead to more efficient sales of goods produced by company PLEAS a.s. Herein, the theoretical knowledge is applied to a real situaution of the marketing research and the proposals resulting thereof are then analyzed and presented.
|
guest ::
RSS feed.