Search

Since 1999, Estonia has conducted the National Forest Inventory (NFI) on the basis of sample plots. This paper presents a new module, incorporating remote-sensing feature variables from airborne laser scanning (ALS) and from multispectral satellite images, for the construction of maps of forest height, standing-wood volume, and tree species composition for the entire country. The models for sparse ALS point clouds yield coefficients of determination of 89.5–94.8% for stand height and 84.2–91.7% for wood volume. For the tree species prediction, the models yield Cohen's kappa values (taking 95% confidence intervals) of 0.69–0.72 upon comparing model results against a previous map, and values of 0.51–0.54 upon comparing model results against NFI sample plots. This paper additionally examines the influence of foliage phenology on the predictions and discusses options for further enhancement of the system.

Masinõppemudelid eeldavad, et andmed pärinevad statsionaarsest jaotusest.Praktikas on tihti vaja mudelitega tõlgendada andmeid, mis pärinevad kiiresti dünaamiliselt muutuvast andmevoost. Seda muutust õppe- ja testvalimis nimetatakse kontseptuaalseks triiviks (ingl k concept drift). Kontseptuaalse triivi olemasolu võib kahjustada mudelennustuste täpsust ja usaldusväärsust. Seetõttu on kontseptuaalse triivi arvestamine väga oluline, et vähendada selle negatiivset mõju tulemustele. Kontseptuaalse triivi arvestamiseks tuleb see kõigepealt tuvastada. Selle tuvastamiseks kasutatakse triivi detektoreid. Reaktiivsed kontseptuaalse triivi detektorid püüavad tuvastada triivi niipe kui see ilmneb, jälgides aluseks oleva masinõppe mudeli toimimist. Tõlgendatavus on masinõppes tähtis ja meetod võib osutuda kasulikuks mitte ainult triivi olemasolu tuvastamiseks andmekogumis, vaid ka triivi põhjuste tuvastamisel ja analüüsimisel.Käesolevas töös rõhutatakse tõlgendatavuse tähtsust triivi tuvastamisel ja esitatakse statistilise triivi tuvastamise meetod (SDDM), mis tuvastab triivi kiiresti arenevates andmevoogudes, kusjuures võrdluses kaasaegsete meetoditega esineb vähem valepositiivseid ja valenegatiivsed tulemusi. Meetod annab ka kontseptuaalse triivi põhjuste tõlgenduse. Töös näidatakse meetodi tõhusust, rakendades seda nii sünteetilistele kui ka reaalsetele andmekogumitele., Machine learning models assume that data is drawn from a stationary distribution. However, in practice, challenges are imposed on models that need to make sense of fast-evolving data streams, where the content of data is changing and evolving dynamically over time. This change between the underlying distributions of the training and test datasets is called concept drift. The presence of concept drift may compromise the accuracy and reliability of prospective computational predictions. Therefore, handling concept drift is of great importance in the direction of diminishing its negative effects on a model's performance. In order to handle concept drift, one has to detect it first. Concept drift detectors have been used to accomplish this - reactive concept drift detectors try to detect drift as soon as it occurs by monitoring the performance of the underlying machine learning model. However, the importance of interpretability in machine learning indicates that it may prove useful to not only detect that drift is occurring in the data, but to also identify and analyze the causes of the drift. In this thesis, the importance of interpretability in drift detection is highlighted and the Statistical Drift Detection Method (SDDM) is presented, which detects drifts in fast-evolving data streams with a smaller number of false positives and false negatives when compared to the state-of-the-art, and has the ability to interpret the cause of the concept drift. The effectiveness of the method is demonstrated by applying it on both synthetic and real-world datasets.

Mobiilioperaatorid koguvad andmeid oma klientide tegevuse kohta mobiilivõrgus. Igal võrgusündmusel on viide mastile, mille külge oli seade parasjagu ühendatud. Teades mastide kattealasid, saab hinnata inimeste trajektoore läbi terve päeva. Nende trajektooride halva ruumilise lahutuse ja ajalise hõreduse tõttu on neist kasuliku informatsiooni eraldamine väljakutsuv ülesanne, mis vajab spetsiifilisi algoritmilisi lahendusi. Trajektooridest paigalseisude tuvastamine on aluseks mitmetele küsimustele lahenduse leidmisele. Selles töös on uuritud ühte paigalseisude leidmise algoritmi ning on tuvastatud mõned selle puudujäägid. Nende puuduste parandamisel muutus algoritm märksa täpsemaks ja töökindlamaks.Lisaks on uuritud trajektooride perioodilisuse kaasamise mõju mudelite asukohahinnangutele., Mobile operators collect data about their clients' activity in the mobile network. Each event made in the mobile network has reference to the antenna the mobile device was connected to at that time. By knowing the coverage areas of the antennas the peoples' trajectories throughout the day can be approximated. Its spatial coarseness and temporal sparseness makes extracting information from this data a compelling task requiring specially crafted algorithmic tools. Detecting when and where did the mobile device stopped is a crucial step that serves as a basis for subsequent data analysis tasks on this data. Here a state of the art stop detection algorithm is analysed and some shortcomings of it identified. The proposed improvements to these have been shown to improve the performance of the stop detection algorithm significantly. Additionally, the possibility of improving the location accuracy by incorporating periodicity into the movement models is investigated.