Global ETD Search
Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.
Results
Showing 1 to 11 of 11 for “"Rare Features"”.
-
Examination and utilization of rare features in text classification of injury narratives
… proposed "Type M+S Word Grouping Method" groups rare and unseen words morphologically and semantically automatically using unlabeled data. Several experiments were conducted for evaluating the grouping effect for three classifiers (MNB, SVM, LR) in three train-test scenarios (1:9, 1:1, 9:1) on …
-
Mining Rare Features in Fingerprints using Core points and Triplet-based Features
… and statistical approach to discover rare features in fingerprints which provides scientific validation for both ten-print and latent fingerprint evidence. A feature is considered rare if it is statistically uncommon; that is, the rare feature should be unique among N (N>100) randomly …
-
Logistic Regression and Cox Hazard Modeling with Sparse High Dimensional Data via Elastic Net Regularization and Graph-Guided Aggregation
Rare features are predictor variables with excessively low rates of nonzeros. It is not uncommon to encounter rare features in settings where data is quantified through one-hot encoding, such as text mining data or genomic data. Rare features pose problems for classic regression techniques due to …
-
Statistical Learning for Structural Patterns with Trees
… whenever another is set to zero; and (2) many features that are counts of rarely occurring events are present, and appropriate aggregation of the rare features may lead to better estimation. In either case, the methods under consideration use a tree or a directed acyclic graph (DAG) that …
-
Minutiae Triplet-based Features with Extended Ridge Information for Determining Sufficiency in Fingerprints
… assessment, and (2) to discover statistically rare features or patterns in fingerprints. These features would help establish an objective minimum-quality baseline for latent prints as well as aid in the latent examination process in making a matching comparison. The proposed methodologies use …
-
Predicting the Emotional Intensity of Tweets
… network Twitter. In this study, we use lexical features, sentiment and emotion lexicons to extract features from tweets, messages of 280 characters or less shared on Twitter. We also use a form of transfer learning – word and sentence embeddings extracted from neural networks trained on large …
-
'Les cent nouvelles nouvelles': A linguistic study of MS Glasgow Hunter 252
… within the document (dialectalisms, archaisms, rare features, cultural references, etc.), through a comprehensive survey of phonology, morphology, syntax and vocabulary. •Expanding and reassessing existing theories on orthographic standardisation and dialectal input in written and, more …
-
Language and the negotiation of identity and belonging in Harub, Saudi Arabia
… The analysis shows that salient linguistic features used in Harūb have become enregistered with Badu identity in terms of ideologies of linguistic differentiation. The historical and social processes of isolation, modernization and marginalization have given rise to discursive practices of …
-
An investigation of Nova Persei (1901) and its environment
… 1901) has been investigated using IRAS far-infrared and optical observations, coupled with detailed modelling of the optical light echoes. Measurements of the time-dependent expansion of the nova ejecta show that significant deceleration is occurring in the SW quadrant, otherwise the expansion …
-
High-resolution spectrographic studies of strange aurora over Svalbard.
… some of these particularly strange and rare features. Within the ARCTICS group I have also been a core contributor to the creation of the Aurora Field Guide and Handbook for Citizen Science, an open-access pair of documents laying out the basic principles of collaboration in aurora …
-
Environmental sedimentology of the lower Middle Ordovician in the lay school section Hogskin Valley, Grainger County, Tennessee
… for environmentaily relevant fossil and sediment features selected from taphonomic, paleontologic, and sedimentologic studies. Presence/absence information was collected on 51 such features or parameters and 255 samples were subjected to Q-mode multivariate cluster analysis to reveal groups of …