ABSTRACT
Users on social networks such as Twitter interact with each other without much knowledge of the real-identity behind the accounts they interact with. This anonymity has created a perfect environment for bot accounts to influence the network by mimicking real-user behaviour. Although not all bot accounts have malicious intent, identifying bot accounts in general is an important and difficult task. In the literature there are three distinct types of feature sets one could use for building machine learning models for classifying bot accounts. These feature-sets are: user profile metadata, natural language features (NLP) extracted from user tweets and finally features extracted from the the underlying social network. Profile metadata and NLP features are typically explored in detail in the bot-detection literature. At the same time less attention has been given to the predictive power of features that can be extracted from the underlying network structure. To fill this gap we explore and compare two classes of embedding algorithms that can be used to take advantage of information that network structure provides. The first class are classical embedding techniques, which focus on learning proximity information. The second class are structural embedding algorithms, which capture the local structure of node neighbourhood. We show that features created using structural embeddings have higher predictive power when it comes to bot detection. This supports the hypothesis that the local social network formed around bot accounts on Twitter contains valuable information that can be used to identify bot accounts.
ABSTRACT
Hyperspectral imaging is a promising clinical imaging modality with multiple applications in wound care, dermatology, and ophthalmology. However, with current technologies, hyperspectral imagers are relatively large and expensive devices, mainly affordable only by hospitals. Multispectral imaging can be a cost-effective alternative for hyperspectral imaging and is capable of bringing diagnostics to primary health care. Multispectral imaging uses known features of tissue chromophores to simplify imaging device design. However, to maintain design simple and cost-effective the number of illumination bands should be minimal. Thus, proper band selection is very important. The goal of the current study is to develop an analytical model for the optimization of band selection for multispectral and narrow-band imaging techniques (e.g., narrow-band microscopy). METHODS: The contrast ratio has been proposed for quantification of image quality of subsurface inhomogeneities in the skin. Based on the two-flux Kubelka-Munk model, we developed an analytical approach which links the contrast ratio with optical tissue parameters. RESULTS: We obtained an explicit analytical solution for the dependence of maximal contrast ratio on optical tissue parameters. Then, we linked the minimally observable contrast ratio (cmin) with the bit depth of the camera, d: cmin = 1/(2d-1). Based on this analysis we were able to derive an explicit expression, which links camera properties with the minimally detectable changes in optical tissue parameters (both scattering and absorption). CONCLUSIONS: The proposed analytical model can be used for rapid assessment and optimization of multispectral and narrow band imaging techniques and for estimation of the accuracy of imaging techniques. The developed model confirms the utility of the contrast ratio for tissue imaging.
