Search results for "neuroverkot"

Application of artificial neural network and genetic algorithm to forecasting of wind power output

2007

DL_Track : Automated analysis of muscle architecture from B-mode ultrasonography images using deep learning

2023

B-mode ultrasound is commonly used to image musculoskeletal tissues, but one major bottleneck is data analysis. Manual analysis is commonly deployed for assessment of muscle thickness, pennation angle and fascicle length in muscle ultrasonography images. However, manual analysis is somewhat subjective, laborious and requires thorough experience. We provide an openly available algorithm (DL_Track) to automatically analyze muscle architectural parameters in ultrasonography images or videos of human lower limb muscles.
 We trained two different neural networks (classic U-net [Ronneberger et al., 2021] and U-net with VGG16 [Simonyan & Zisserman, 2015] pretrained encoder) one to detect …

Identifying the Sales Patterns of Online Stores with Self-Organising Maps on Time Series Data

2018

Electronic commerce, especially in the business-to-consumer (B2C) context, has for years been a popular research topic in information systems (IS). However, the prior research on the topic has traditionally been dominated by the consumer focus instead of the business focus of online stores. For example, whereas various segmentations exist for online consumers based on their purchase behaviour, no such segmentations have been developed for online stores based on their sales patterns. In this study, our objective is to address this gap in prior research by identifying the most typical sales patterns of online stores operating in the B2C context. By using self-organising maps (SOM) to analyse …

Neuroverkkojen regularisointimenetelmät

2016

Ylisovitus on yleinen ongelma ohjatussa oppimisessa, missä malli oppii suoriutumaan hyvin oppimisessa käytetyllä datalla, mutta alisuoriutuu oppimisen aikana näkemättömällä datalla. Regularisointimenetelmillä pyritään vähentämään ylisovitusta ohjatun oppimisen sovellutuksissa. Tämä tutkielma keskittyy tutkimaan ja kartoittamaan erilaisia neuroverkoissa käytettyjä regularisointimenetelmiä. Overfitting is a common problem in supervised learning, where a model learns to perform well with the data used to train it, but underperforms with data it has not seen during the training. Regularization methods are used to reduce overfitting in applications of supervised learning. This paper focuses on r…