FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2891433624> ?p ?o ?g. }

• W2891433624 abstract "In 2015, 2.5 quintillion bytes of data were daily generated worldwide of which 90% were unstructured data that do not follow any pre-defined model. These data can be found in a great variety of formats among them are texts, images, audio tracks, or videos. With appropriate techniques, this massive amount of data is a goldmine from which one can extract a variety of meaningful embedded information. Among those techniques, machine learning algorithms allow multiple processing possibilities from compact data representation, to data clustering, classification, analysis, and synthesis, to the detection of outliers. Data modeling is the first step for performing any of these tasks and the accuracy and reliability of this initial step is thus crucial for subsequently building up a complete data processing framework. The principal motivation behind my work is the over-use of the Gaussian assumption for data modeling in the literature. Though this assumption is probably the best to make when no information about the data to be modeled is available, in most cases studying a few data properties would make other distributions a better assumption. In this thesis, I focus on proportional data that are most commonly known in the form of histograms and that naturally arise in a number of situations such as in bag-of-words methods. These data are non-Gaussian and their modeling with distributions belonging the Dirichlet family, that have common properties, is expected to be more accurate. The models I focus on are the hidden Markov models, well-known for their capabilities to easily handle dynamic ordered multivariate data. They have been shown to be very effective in numerous fields for various applications for the last 30 years and especially became a corner stone in speech processing. Despite their extensive use in almost all computer vision areas, they are still mainly suited for Gaussian data modeling. I propose here to theoretically derive different approaches for learning and applying to real-world situations hidden Markov models based on mixtures of Dirichlet, generalized Dirichlet, Beta-Liouville distributions, and mixed data. Expectation-Maximization and variational learning approaches are studied and compared over several data sets, specifically for the task of detecting and localizing unusual events. Hybrid HMMs are proposed to model mixed data with the goal of detecting changes in satellite images corrupted by different noises. Finally, several parametric distances for comparing Dirichlet and generalized Dirichlet-based HMMs are proposed and extensively tested for assessing their robustness. My experimental results show situations in which such models are worthy to be used, but also unravel their strength and limitations." @default.
• W2891433624 created "2018-09-27" @default.
• W2891433624 creator A5061523229 @default.
• W2891433624 date "2017-10-15" @default.
• W2891433624 modified "2023-09-27" @default.
• W2891433624 title "Non-Gaussian data modeling with hidden Markov models" @default.
• W2891433624 cites W1528056001 @default.
• W2891433624 cites W1528124672 @default.
• W2891433624 cites W1529958853 @default.
• W2891433624 cites W1548605257 @default.
• W2891433624 cites W1592240776 @default.
• W2891433624 cites W1596052433 @default.
• W2891433624 cites W1613782521 @default.
• W2891433624 cites W1640814127 @default.
• W2891433624 cites W181338166 @default.
• W2891433624 cites W1931450083 @default.
• W2891433624 cites W1947560069 @default.
• W2891433624 cites W1967456674 @default.
• W2891433624 cites W1972302851 @default.
• W2891433624 cites W1973057208 @default.
• W2891433624 cites W1978879525 @default.
• W2891433624 cites W1983249544 @default.
• W2891433624 cites W1983364832 @default.
• W2891433624 cites W1986437065 @default.
• W2891433624 cites W1988627679 @default.
• W2891433624 cites W1989387811 @default.
• W2891433624 cites W1989856502 @default.
• W2891433624 cites W1991251598 @default.
• W2891433624 cites W1993409002 @default.
• W2891433624 cites W1995430360 @default.
• W2891433624 cites W1996970712 @default.
• W2891433624 cites W1997534118 @default.
• W2891433624 cites W2000930721 @default.
• W2891433624 cites W2007321142 @default.
• W2891433624 cites W2010399676 @default.
• W2891433624 cites W2011384856 @default.
• W2891433624 cites W2012185915 @default.
• W2891433624 cites W2013848746 @default.
• W2891433624 cites W2014801693 @default.
• W2891433624 cites W2015461918 @default.
• W2891433624 cites W2025589645 @default.
• W2891433624 cites W2033178790 @default.
• W2891433624 cites W2033493830 @default.
• W2891433624 cites W2033810668 @default.
• W2891433624 cites W2035611878 @default.
• W2891433624 cites W2036073303 @default.
• W2891433624 cites W2036194630 @default.
• W2891433624 cites W2036631917 @default.
• W2891433624 cites W2039344516 @default.
• W2891433624 cites W2046269793 @default.
• W2891433624 cites W2051110770 @default.
• W2891433624 cites W2055337076 @default.
• W2891433624 cites W2056109357 @default.
• W2891433624 cites W2072449446 @default.
• W2891433624 cites W2073753828 @default.
• W2891433624 cites W207560109 @default.
• W2891433624 cites W2075978670 @default.
• W2891433624 cites W2079023123 @default.
• W2891433624 cites W2082697323 @default.
• W2891433624 cites W2086413055 @default.
• W2891433624 cites W2093212899 @default.
• W2891433624 cites W2094952625 @default.
• W2891433624 cites W2096784803 @default.
• W2891433624 cites W2098562545 @default.
• W2891433624 cites W2100807019 @default.
• W2891433624 cites W2105594594 @default.
• W2891433624 cites W2110934250 @default.
• W2891433624 cites W2111284386 @default.
• W2891433624 cites W2113786023 @default.
• W2891433624 cites W2115461049 @default.
• W2891433624 cites W2115597079 @default.
• W2891433624 cites W2115979064 @default.
• W2891433624 cites W2116064335 @default.
• W2891433624 cites W2119605622 @default.
• W2891433624 cites W2119812628 @default.
• W2891433624 cites W2121436925 @default.
• W2891433624 cites W2122361470 @default.
• W2891433624 cites W2124386111 @default.
• W2891433624 cites W2125838338 @default.
• W2891433624 cites W2128730107 @default.
• W2891433624 cites W2130117172 @default.
• W2891433624 cites W2131550388 @default.
• W2891433624 cites W2134392507 @default.
• W2891433624 cites W2136677996 @default.
• W2891433624 cites W2139688603 @default.
• W2891433624 cites W2140170007 @default.
• W2891433624 cites W2140233853 @default.
• W2891433624 cites W2142412278 @default.
• W2891433624 cites W2142635246 @default.
• W2891433624 cites W2143143555 @default.
• W2891433624 cites W2144301353 @default.
• W2891433624 cites W2146916348 @default.
• W2891433624 cites W2147850603 @default.
• W2891433624 cites W2152548630 @default.
• W2891433624 cites W2152594362 @default.
• W2891433624 cites W2153086947 @default.
• W2891433624 cites W2153481163 @default.
• W2891433624 cites W2157932304 @default.
• W2891433624 cites W2163513674 @default.
• W2891433624 cites W2164489414 @default.

• next