Search NASA⌕ Search

SEARCH · Search NASA

Results for “missing values”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 91 records · Page 5

FAIRification, Quality Assessment, and Missingness Pattern Discovery for Spatiotemporal Photovoltaic Data

The ongoing growth of the photovoltaic market has pushed the demand for power forecasting and performance evaluation for a huge population of PV power plants. Through access to a large number of time series data sets from different power plants, we have found common issues that impede the modeling process. Namely, the time series data are hard to transfer between groups due to differences in variable nomenclature, and the quality of the data sets can vary. We address the issue of variable nomenclature by FAIRifying spatiotemporal PV time series data. Through the creation of a solar power plant ontology, we propose standards for the naming and structure of metadata used to describe the data from these power plants. Using the structure from this ontology, we have developed both R and Python packages for the automation of the FAIRification process. We have also developed an R package that automates the analysis of the quality of a data set through the designation of letter grades. With access to large time series data sets across many power plants, we can utilize spatiotemporal coherence between the sites in order to improve the quality of our data. To solve the issue of data missingness, we propose the use of Spatiotemporal-GNN autoencoders to detect and impute missing values from a data set by utilizing data from power plants nearby.

14 SOLAR ENERGY↗

Feature Extraction: Improving Remote Sensor Classification of Non-Proliferation

This research focuses on developing algorithms for nuclear non-proliferation detection using remote sensor modeling. To improve the performance of classification models, we implemented a data pipeline with feature extraction. This pipeline takes raw data and transforms it into smaller data points called features that still describe the model. Improving this classification works towards the departments of energy’s missions of ensuring American’s security and prosperity by creating technology that addresses nuclear challenges. To conduct this analysis, we used the Python programming language and some key packages, including tsfresh and TSFEL. Originally tsfresh was selected because it has the most statistical features out of all the packages. Later TSFEL was incorporated due to the additional features it can extract from data, such as temporal and spectral. However, feature extraction becomes challenging in the presence of missing values. In this case, two additional Python packages were added to our workflow, NumPy and pandas, allowing for the feature extraction process to handle unknown values. Our data pipeline was tested on data collected from a simulation that describes the process state of a physical example. The results show the pipeline’s capability to consume and extract a total 17 features from tabular data. Future work includes producing classifications using decision tree-based models such as XGBoost and improving data collection by analyzing feature importance.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Jobs, jobs, jobs: what’s an analyst to do?

Analysts and economists often face the task of using employment metrics to characterize industries of interest. Some key challenges can be understanding where to find employment metrics, the differences in various employment metrics, and when each metric should be used. This article analyzes a variety of publicly available employment data for the United States and compares these data. A detailed description of the intricacies of each data source is provided, which covers factors such as regionality, industry breakout, periodicity, and the types of jobs included. This article provides several case study examples, using the oil and gas extraction, coal mining, and chemical manufacturing sectors to portray challenges data users may face when developing employment estimates that suit their needs. Data users should be aware of a variety of data sources to understand alternative analysis options when data limitations are present and to determine which data source best meets their needs. Instances may occur in which information from one dataset may be used to help impute missing values.

99 GENERAL AND MISCELLANEOUS↗

An Updated Global Grid Point Surface Air Temperature Anomaly Data Set: 1851-1990 (revised 1991) (NDP-020)

This NDP presents land-based monthly surface-air-temperature anomalies (departures from a 1951-1970 reference period mean) on a 5° latitude by 10° longitude global grid. Monthly surface-air-temperature anomalies (departures from a 1957-1975 reference period mean) for the Antarctic (grid points from 65°S to 85°S) are presented in a similar way as a separate data set. The data were derived primarily from the World Weather Records and from the archives of the United Kingdom Meteorological Office. This long-term record of temperature anomalies may be used in studies addressing possible greenhouse-gas-induced climate changes. To date, the data have been employed in producing regional, hemispheric, and global time series for determining whether recent (i.e., post-1900) warming trends have taken place. The present updated version of this data set is identical to the earlier version for all records from 1851-1978 except for the addition of the Antarctic surface-air-temperature anomalies beginning in 1957. Beginning with the 1979 data, this package differs from the earlier version in several ways. Erroneous data for some sites have been corrected after a review of the actual station temperature data, and inconsistencies in the representation of missing values have been removed. For some grid locations, data have been added from stations that had not contributed to the original set. Data from satellites have also been used to correct station records where large discrepancies were evident. The present package also extends the record by adding monthly surface-air-temperature anomalies for the Northern (grid points from 85°N to 0°) and Southern (grid points from 5°S to 60°S) Hemispheres for 1985-1990. In addition, this updated package presents the monthly-mean-temperature records for the individual stations that were used to produce the set of gridded anomalies. The periods of record vary by station. Northern Hemisphere data have been corrected for inhomogeneities, while Southern Hemisphere data are presented in uncorrected form.For access to the data files, click this link to the CDIAC data transition website: http://cdiac.ess-dive.lbl.gov/ndps/ndp020.html

54 ENVIRONMENTAL SCIENCES↗

STAIR 2.0: A Generic and Automatic Algorithm to Fuse Modis, Landsat, and Sentinel-2 to Generate 10 m, Daily, and Cloud-/Gap-Free Surface Reflectance Product

Remote sensing datasets with both high spatial and high temporal resolution are critical for monitoring and modeling the dynamics of land surfaces. However, no current satellite sensor could simultaneously achieve both high spatial resolution and high revisiting frequency. Therefore, the integration of different sources of satellite data to produce a fusion product has become a popular solution to address this challenge. Many methods have been proposed to generate synthetic images with rich spatial details and high temporal frequency by combining two types of satellite datasets—usually frequent coarse-resolution images (e.g., MODIS) and sparse fine-resolution images (e.g., Landsat). In this paper, we introduce STAIR 2.0, a new fusion method that extends the previous STAIR fusion framework, to fuse three types of satellite datasets, including MODIS, Landsat, and Sentinel-2. In STAIR 2.0, input images are first processed to impute missing-value pixels that are due to clouds or sensor mechanical issues using a gap-filling algorithm. The multiple refined time series are then integrated stepwisely, from coarse- to fine- and high-resolution, ultimately providing a synthetic daily, high-resolution surface reflectance observations. We applied STAIR 2.0 to generate a 10-m, daily, cloud-/gap-free time series that covers the 2017 growing season of Saunders County, Nebraska. Moreover, the framework is generic and can be extended to integrate more types of satellite data sources, further improving the quality of the fusion product. View Full-Text

47 OTHER INSTRUMENTATION↗

Quality Control Methods for Advanced Metering Infrastructure Data

While urban-scale building energy modeling is becoming increasingly common, it currently lacks standards, guidelines, or empirical validation against measured data. Empirical validation necessary to enable best practices is becoming increasingly tractable. The growing prevalence of advanced metering infrastructure has led to significant data regarding the energy consumption within individual buildings, but is something utilities and countries are still struggling to analyze and use wisely. In partnership with the Electric Power Board of Chattanooga, Tennessee, a crude OpenStudio/EnergyPlus model of over 178,000 buildings has been created and used to compare simulated energy against actual, 15-min, whole-building electrical consumption of each building. In this study, classifying building type is treated as a use case for quantifying performance associated with smart meter data. This article attempts to provide guidance for working with advanced metering infrastructure for buildings related to: quality control, pathological data classifications, statistical metrics on performance, a methodology for classifying building types, and assess accuracy. Advanced metering infrastructure was used to collect whole-building electricity consumption for 178,333 buildings, define equations for common data issues (missing values, zeros, and spiking), propose a new method for assigning building type, and empirically validate gaps between real buildings and existing prototypes using industry-standard accuracy metrics.

24 POWER TRANSMISSION AND DISTRIBUTION↗

A Photochemical Phosphorus-Hydrogen-Oxygen Network for Hydrogen-dominated Exoplanet Atmospheres

Due to the detection of phosphine (PH 3 ) in the solar system gas giants Jupiter and Saturn, PH 3 has long been suggested to be detectable in exosolar substellar atmospheres too. However, to date, direct detection of phosphine has proven to be elusive in exoplanet atmosphere surveys. We construct an updated phosphorus-hydrogen-oxygen (PHO) photochemical network suitable for the simulation of gas giant hydrogen-dominated atmospheres. Using this network, we examine PHO photochemistry in hot Jupiter and warm Neptune exoplanet atmospheres at solar and enriched metallicities. Our results show for HD 189733b-like hot Jupiters that HOPO, PO, and P 2 are typically the dominant P carriers at pressures important for transit and emission spectra, rather than PH 3 . For GJ1214b-like warm Neptune atmospheres our results suggest that at solar metallicity PH 3 is dominant in the absence of photochemistry, but is generally not in high abundance for all other chemical environments. At 10 and 100 times solar, small oxygenated phosphorus molecules such as HOPO and PO dominate for both thermochemical and photochemical simulations. The network is able to reproduce well the observed PH 3 abundances on Jupiter and Saturn. Despite progress in improving the accuracy of the PHO network, large portions of the reaction rate data remain with approximate, uncertain, or missing values, which could change the conclusions of the current study significantly. Improving understanding of the kinetics of phosphorus-bearing chemical reactions will be a key undertaking for astronomers aiming to detect phosphine and other phosphorus species in both rocky and gaseous exoplanetary atmospheres in the near future.

atmospheric composition↗

NGEE Arctic Plant Traits: Shrub Transects, Kougarok Road Mile Marker 64, Seward Peninsula, Alaska, 2016-2017

Transect data is from point center quarter (PCQ) surveys for shrub density performed in July 2016 at the Kougarok site and in July 2017 at the Kougarok, Teller, and Council sites. For each sample point along the transects, moving averages for shrub density and shrub basal area are provided along with GPS coordinates. Average shrub height and active layer depth are recorded for each sample point as well. Target shrub species differ across the transects: some surveys targeted a mixture of tall shrub species while others targeted specific species. In 2017 insect damage was also quantified. The dataset contains two *.csv data files and one *.pdf. Missing Values are represented by "NA". The Next-Generation Ecosystem Experiments: Arctic (NGEE Arctic), was a research effort to reduce uncertainty in Earth System Models by developing a predictive understanding of carbon-rich Arctic ecosystems and feedbacks to climate. NGEE Arctic was supported by the Department of Energy's Office of Biological and Environmental Research. The NGEE Arctic project had two field research sites: 1) located within the Arctic polygonal tundra coastal region on the Barrow Environmental Observatory (BEO) and the North Slope near Utqiagvik (Barrow), Alaska and 2) multiple areas on the discontinuous permafrost region of the Seward Peninsula north of Nome, Alaska. Through observations, experiments, and synthesis with existing datasets, NGEE Arctic provided an enhanced knowledge base for multi-scale modeling and contributed to improved process representation at global pan-Arctic scales within the Department of Energy's Earth system Model (the Energy Exascale Earth System Model, or E3SM), and specifically within the E3SM Land Model component (ELM).

54 ENVIRONMENTAL SCIENCES↗

Utility-scale Building Type Assignment Using Smart Meter Data

United States building energy use accounted for 40% of total energy use, 74% of peak demand, and $412 billion in 2019. Building energy modeling allows researchers to simulate building physics, gain insights into possible energy/demand saving opportunities, and assess cost-effective resilience amidst climate change. Many building features needed to create building energy models are readily available such as 2D footprints and LiDAR (height). A critical feature that is not generally obtainable is the building type. In partnership with a utility, a years worth of real-world, 15-minute electrical use data has been examined. The smart meter data is compared to 97 different prototype building energy models to assign building type. Real-world considerations including data preparation, quality assurance, and handling of missing values for advanced metering infrastructure data are addressed. Euclidean distance for pattern-matching of energy use, dynamic time warping, and time-window statistics with machine learning are compared for determining building type from measured electricity use.

Bass, Brett↗

A Data Quality-Aware Framework to Reliably Forecast Photovoltaic Generation and Consumer Load for an Improved Resilience of Microgrids

Photovoltaic (PV) power and consumer load forecasting plays a critical role to ensure operational resilience of the electric grid. Most data-driven forecasting algorithms rely heavily on the continuous availability of good quality data for periodic training and validation. When deployed at the grid’s edge, prolonged disruptions to communications during extreme events degrade data quality. Factors such as missing observations, epistemic uncertainties, data drift, and concept drift are manifestations of data quality that impact the generalization of such field-deployed forecasting models. Currently, there exists no mechanism in the literature to dynamically switch between models under varying degrees of data quality as quantified by certain metrics for each factor highlighted above. This paper addresses this shortcoming by conceptually introducing a data qualityaware framework for reliable PV generation and consumer load forecasting. The framework’s design incorporates components of missing values, divergence tests, and continuous monitoring of generalization performance to detect changes in data quality caused by communications disruptions and trigger specific classes of forecasting models grouped under three use cases (UC1- UC3). As a first step towards validating this framework, real data collected from an actual field microgrid system is used to demonstrate the viability of the three use cases. Results show that the performance is the best in UC1 with an unadjusted R-square value of 0.954, followed by 0.939 for UC2 and 0.757 for UC3.

Sundararajan, Aditya↗

Combined Statistical Analyses of Peptide Intensities and Peptide Occurrences Improves Identification of Significant Peptides from MS-based Proteomics Data

Liquid chromatography-mass spectrometry-based (LC-MS) proteomics uses peak intensities of proteolytic peptides to infer the differential abundance of peptides/proteins. However, substantial run-to-run variability in peptide intensities and observations (presence/absence) of peptides makes data analysis quite challenging. The missing abundance values in LC-MS proteomics data are difficult to address with traditional imputation-based approaches because the mechanisms by which data are missing are unknown a priori. Data can be missing due to random mechanisms such as experimental error, or non-random mechanisms such as a true biological effect. We present a statistical approach that uses a test of independence known as a G-test to test the null hypothesis of independence between the number of missing values and the experimental groups. We pair the G-test results evaluating independence of missing data (IMD) with a standard analysis of variance (ANOVA) that uses only means and variances computed from the observed data. Each peptide is therefore represented by two statistical confidence metrics, one for qualitative differential observation and one for quantitative differential intensity. We use two simulated and two real LC-MS datasets to demonstrate the robustness and sensitivity of the ANOVA-IMD approach for assigning confidence to peptides with significant differential abundance among experimental groups.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Scalable tensor factorizations with incomplete data.

The problem of incomplete data - i.e., data with missing or unknown values - in multi-way arrays is ubiquitous in biomedical signal processing, network traffic analysis, bibliometrics, social network analysis, chemometrics, computer vision, communication networks, etc. We consider the problem of how to factorize data sets with missing values with the goal of capturing the underlying latent structure of the data and possibly reconstructing missing values (i.e., tensor completion). We focus on one of the most well-known tensor factorizations that captures multi-linear structure, CANDECOMP/PARAFAC (CP). In the presence of missing data, CP can be formulated as a weighted least squares problem that models only the known entries. We develop an algorithm called CP-WOPT (CP Weighted OPTimization) that uses a first-order optimization approach to solve the weighted least squares problem. Based on extensive numerical experiments, our algorithm is shown to successfully factorize tensors with noise and up to 99% missing data. A unique aspect of our approach is that it scales to sparse large-scale data, e.g., 1000 x 1000 x 1000 with five million known entries (0.5% dense). We further demonstrate the usefulness of CP-WOPT on two real-world applications: a novel EEG (electroencephalogram) application where missing data is frequently encountered due to disconnections of electrodes and the problem of modeling computer network traffic where data may be absent due to the expense of the data collection process.

99 GENERAL AND MISCELLANEOUS↗

Observed and Imputed Volumetric Soil Water Content Timeseries for the New Mexico Elevation Gradient

Reliable soil water content (SWC) data are essential for understanding dryland ecosystem dynamics, but high-frequency SWC sensors often fail, creating gaps in critical datasets. To address this, we developed a Bayesian mixture model that imputes missing SWC using both linear interpolation and an ecosystem water balance model (SOILWAT2), tested across six AmeriFlux eddy covariance tower sites in the New Mexico Elevation Gradient, demonstrating its effectiveness in reconstructing SWC patterns while providing insights into the factors driving SWC variability. Daily volumetric soil water content (SWC) data are provided as csv-formatted spreadsheets for the six AmeriFlux sites (US-Seg, US-Ses, US-Wjs, US-Mpi, US-Vcp, and US-Vcs). For each site there is an observed SWC file (site_SWC_gapfill.csv) and a file that contains imputed SWC (imputed_SWC_site.csv). The observed SWC files contain temperature corrected sensor values, tower precipitation data, as well as outputs from SOILWAT2 simulations that were used to impute SWC. The imputed files contain the original observed SWC values and the imputed missing SWC values. When SWC was missing from the original data, the missing value was imputed based on the Bayesian imputation mixture model. The posterior mean of all imputed values is reported as "mean_X". When the observed SWC was NOT missing, mean_X = observed SWC value (original data). The standard deviation, 2.5th percentile and the 97.5th percentile for the imputed values are also reported in the imputed files. There are readme text files for each file type explaining the contents of each column.

54 ENVIRONMENTAL SCIENCES↗

Search for New Physics in a Final State with Same-Sign Dileptons, Jets, and Missing Transverse Energy at 7 TeV Center of Mass Energy

We report on a search for new physics in a final state with two same sign leptons, missing transverse energy, and significant hadronic activity at a center of mass energy $\sqrt{s}$ = 7 TeV. The data were collected with the CMS detector at the CERN LHC and correspond to an integrated luminosity of 0.98 $\mathrm{fb}^{-1}$. Data--driven methods are developed to estimate the dominant Standard Model backgrounds. No evidence for new physics is observed. The dominant background to the analysis comes from failures of lepton identification in Standard Model $t\bar{t}$ events. The $t\bar{t}$ production cross section in the dilepton final state is measured using 3.1 $\mathrm{fb}^{-1}$ of data. The cross section is measured to be 194 $\pm$ 72 (stat) $\pm$ 24 (syst) $\pm$ 21 (lumi) pb. An algorithm is developed that uses tracking information to improve the reconstruction of missing transverse energy. The reconstruction of missing transverse energy is commissioned using the first collisio ns recorded at 0.9, 2.36 and 7 TeV data. Events with abnormally large values of missing transverse energy are identified as arising from anomalous signals in the calorimeters. Tools are developed to identify and remove these anomalous signals.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Radiation Doses to Members of the U.S. Population from Ubiquitous Radionuclides in the Body: Part 2, Methods and Dose Calculations

This paper is part two of a three-part series investigating annual effective doses to residents of the United States from intakes of ubiquitous radionuclides, including radionuclides occurring naturally, radionuclides whose concentrations are technologically enhanced, and anthropogenic radionuclides. This series of papers explicitly excludes intakes from inhaling 222Rn, 220Rn, and their short-lived decay products; it also excludes intakes of radionuclides in occupational and medical settings. Part one reviewed, summarized, characterized, and grouped all published and some unpublished data for U.S. residents on ubiquitous radionuclide concentrations in tissues and organs. Assumptions about equilibrium with long-lived parents are made for the 28 other radionuclides in these series lacking data. This paper describes the methods developed to group the collected data into source regions described in the Radiation Dose Assessment Resource (RADAR) dosimetric methodology. Methods for converting the various units of data published over 50 years into a standard form are developed and described. Often, meaningful values of uncertainty of measurements were not published so that variability in data sets is confounded with measurement uncertainty. A description of the methods developed to estimate variability is included in this paper. The data described in part one are grouped by gender and age to match the RADAR dosimetric phantoms. Within these phantoms, concentration values are grouped into source tissue regions by radionuclide, and they are imputed for source regions lacking tissue data. Radionuclide concentrations are then imputed for other phantoms’ source regions with missing concentration values, and the uncertainties of the imputed values are increased. The content concentrations of hollow organs are calculated, and activities are apportioned to the bone source regions using assumptions about each radionuclide’s bone-seeking behavior. The data sets are then ready to be used to estimate equivalent doses to target tissues from these source regions. The target tissues are then mapped to lists of tissues with International Commission on Radiation Protection (ICRP) tissue weighting factors, or they are mapped to surrogate tissue regions when there is no direct match. Effective doses, using ICRP tissue weighting factors recommended in 1977, 1990, and 2007, are calculated from tissue and organ equivalent doses.

59 BASIC BIOLOGICAL SCIENCES↗

Automated Demand Response: The Missing Link in the Electricity Value Chain

In 2006, the Public Interest Energy Research Program (PIER) Demand Response Research Center (DRRC) at Lawrence Berkeley National Laboratory initiated research into Automated Demand Response (OpenADR) applications in California industry. The goal is to improve electric grid reliability and lower electricity use during periods of peak demand. The purpose of this research is to begin to define the relationship among a portfolio of actions that industrial facilities can undertake relative to their electricity use. This ?electricity value chain? defines energy management and demand response (DR) at six levels of service, distinguished by the magnitude, type, and rapidity of response. One element in the electricity supply chain is OpenADR, an open-standards based communications system to send signals to customers to allow them to manage their electric demand in response to supply conditions, such as prices or reliability, through a set of standard, open communications. Initial DRRC research suggests that industrial facilities that have undertaken energy efficiency measures are probably more, not less, likely to initiate other actions within this value chain such as daily load management and demand response. Moreover, OpenADR appears to afford some facilities the opportunity to develop the supporting control structure and to"demo" potential reductions in energy use that can later be applied to either more effective load management or a permanent reduction in use via energy efficiency. Under the right conditions, some types of industrial facilities can shift or shed loads, without any, or minimal disruption to operations, to protect their energy supply reliability and to take advantage of financial incentives.1 In 2007 and 2008, 35 industrial facilities agreed to implement OpenADR, representing a total capacity of nearly 40 MW. This paper describes how integrated or centralized demand management and system-level network controls are linked to OpenADR systems. Case studies of refrigerated warehouses and wastewater treatment facilities are used to illustrate OpenADR load reduction potential. Typical shed and shift strategies include: turning off or operating compressors, aerator blowers and pumps at reduced capacity, increasing temperature set-points or pre-cooling cold storage areas and over-oxygenating stored wastewater prior to a DR event. This study concludes that understanding industrial end-use processes and control capabilities is a key to support reduced service during DR events and these capabilities, if DR enabled, hold significant promise in reducing the electricity demand of the industrial sector during utility peak periods.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗

Overview of the FAF3 Freight Flow Matrix Construction Process

PROJECT DESCRIPTION The FAF3 is a FHWA freight data product which provides a national O-D matrix of commodity flows to, from, and within the United States. FAF3 freight flows are reported in terms of both annual tons and annual dollars of freight moved by mode of transportation. Based largely on the 2007 CFS, FAF3 utilizes domestic freight flow characteristics, geographic regions, and the SCTG commodity coding system from CFS. However, many freight flows were not captured by the 2007 CFS due to scope and sample size limitations. Approximately 100,000 establishments were sampled out of some 754,000 freight moving establishments in 2007 and imports are out of scope entirely. To estimate missing data values, the approach taken in FAF3 was to use a combination of a novel Log-linear modeling approach (LLM) with an iterative proportional fitting (IPF) routine that also uses additional data inputs to fill in the missing pieces. The complete FAF3 O-D Commodity Mode database is made up of 131 Origins x 131 Destinations x 43 Commodity Classes x 8 Modal categories, for annual tons and dollars. This poster illustrates how the 2007 CFS data were integrated with several additional data sources using LLM and IPF to create a comprehensive FAF3 national freight flow matrix. More detailed documentation on the sources and methods utilized in the development of FAF3 are available from the FHWA website at the following website: http://www.ops.fhwa.dot. gov/freight/freight_analysis/faf/index.htm.

99 GENERAL AND MISCELLANEOUS↗

Expectation-maximization algorithms for learning a finite mixture of univariate survival time distributions from partially specified class values

Heterogeneity exists on a data set when samples from di erent classes are merged into the data set. Finite mixture models can be used to represent a survival time distribution on heterogeneous patient group by the proportions of each class and by the survival time distribution within each class as well. The heterogeneous data set cannot be explicitly decomposed to homogeneous subgroups unless all the samples are precisely labeled by their origin classes; such impossibility of decomposition is a barrier to overcome for estimating nite mixture models. The expectation-maximization (EM) algorithm has been used to obtain maximum likelihood estimates of nite mixture models by soft-decomposition of heterogeneous samples without labels for a subset or the entire set of data. In medical surveillance databases we can find partially labeled data, that is, while not completely unlabeled there is only imprecise information about class values. In this study we propose new EM algorithms that take advantages of using such partial labels, and thus incorporate more information than traditional EM algorithms. We particularly propose four variants of the EM algorithm named EM-OCML, EM-PCML, EM-HCML and EM-CPCML, each of which assumes a specific mechanism of missing class values. We conducted a simulation study on exponential survival trees with five classes and showed that the advantages of incorporating substantial amount of partially labeled data can be highly signi cant. We also showed model selection based on AIC values fairly works to select the best proposed algorithm on each specific data set. A case study on a real-world data set of gastric cancer provided by Surveillance, Epidemiology and End Results (SEER) program showed a superiority of EM-CPCML to not only the other proposed EM algorithms but also conventional supervised, unsupervised and semi-supervised learning algorithms.

97 MATHEMATICS AND COMPUTING↗