Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Queries”

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 253 records · Page 14

BrickQA: Bridging the Semantic Gap in Building Operations with Dynamic Graph Exploration

While standardized ontologies like the Brick schema address data heterogeneity in Building Automation Systems (BAS), accessing this semantic data remains a challenge as domain experts often lack the expertise to formulate complex SPARQL queries. To bridge this gap, we present BrickQA, a Large Language Model (LLM)-based framework that translates natural language into executable SPARQL queries through structured query decomposition, dynamic schema exploration, and inline validation. BrickQA utilizes an iterative reasoning agent to actively navigate graph topology through dynamic exploration actions without requiring exhaustive context injection or model fine-tuning. This approach effectively mitigates hallucinations, particularly in large-scale building knowledge graphs. Empirical evaluation on BuildingQA, a standardized benchmark, demonstrates that BrickQA significantly outperforms ReAct baselines, delivering a 0.291–0.355 absolute F1 improvement while achieving 3 × –12.7 × higher token cost-efficiency. Beyond these metrics, the framework maintains structural fidelity across heterogeneous buildings and remains resilient to ambiguous queries without requiring site-specific fine-tuning. Furthermore, a case study on operational analytics validates the framework’s capability to handle temporal and aggregation constraints, effectively transforming abstract semantic models into actionable facility management insights.1

Ko, Yun-Dam↗

Spatially Accelerated Winding Numbers for Curved Geometry

The generalized winding number (GWN) is a scalar field that supports robust containment queries on curved geometry, including non-watertight, overlapping, and nested boundary representations. While queries can be easily parallelized over samples, direct evaluation on parametric curves and surfaces remains costly for large and complex models. Fast, state-of-the-art GWN approaches leverage a spatial index to approximate the GWN, typically coupled with a Taylor expansion which approximates the GWN contribution for far clusters of geometric primitives. However, such methods operate only on discrete inputs such as triangle meshes and point clouds, and would introduce containment errors near boundaries if applied to curved input. We extend support for fast GWN evaluation over arbitrary collections of NURBS curves in 2D and trimmed NURBS patches in 3D via a Bounding Volume Hierarchy that stores efficiently precomputed moment data in the hierarchy nodes. When querying the hierarchy, approximations for far clusters are used alongside direct evaluation for nearby NURBS primitives, achieving sub-linear complexity while preserving the geometric features in the vicinity of the query point. Central to our performance improvements is an adaptive subdivision strategy for NURBS primitives during a preprocessing phase, creating better spatial partitions while retaining the same accuracy for containment decisions as a direct evaluation. We demonstrate the performance and accuracy of our approach across a large collection of 2D and 3D datasets.

Computer science↗

Simulation INsight and Analysis

Sina is a tool set for modern scientific data management that provides flexible, light-weight support of non-bulk data capture for retention in and queries against SQL and noSQL data stores. HPC simulations traditionally maintain their data in files. Extracting data of interest for subsequent analysis then requires the time-consuming process of traversing directories and scraping data from files in a variety of formats. Sina facilitates capturing relevant data during execution or post-processing of simulation runs for retention in and queries from a modern data store. The tools are sufficiently general to allow for the inclusion of new fields as scientists learn more about their data. Libraries, currently in C++ and Python, and a command line interface (CLI) are provided. Sina's flexibility starts with a general schema, in JSON, for the collection of non-bulk simulation data. JSON provides a flexible, human-readable representation of the data that of interest. Sina currently has a C++ library for simulations to write data to and read from a schema-compliant file for subsequent ingestion into one of the supported data stores. However, applications are free to write their data directly into a schema-compliant file. Python packages provide data ingestion, management, query, and export capabilities. A command line interface (CLI) provides simplified access to these features. A common application programming interface (API) is used to maintain and query data in any of the supported data stores, which are currently limited to SQL and Apache Cassandra (a column store). Tutorials, demonstrations, and examples illustrate aspects of the process using scripts and Jupyter notebooks.

Pauli, EstebanT↗

A Data Processing Pipeline To Extract A Knowledge Graph From Heterogeneous Data For Socio-technical Analysis Of Critical Infrastructure Influence

The code is written in Python and consists of the following pipeline that is implemented in Apache Airflow. This pipeline intends to understand the companies that are directly or indirectly involved with a type of critical infrastructure system at some point in that system's lifecycle. The pipeline takes a configuration file that specifies a list of initial companies to consider, a geographic region of interest, and a set of SEC form types as well as other data sources (e.g. CrunchBase) from which to extract entities and relations. There are four main components to this pipeline as currently implemented: Entity Extraction, Network Construction, Analysis, and Visualization. First, Entity Extraction, is implemented as the `topear-extract_organizations` Apache Airflow workflow. Given an initial query that specifies a geographic region of interest and a time interval, the software will extract CI facilities of interest and organizations that have a direct influence relationship to those facilities (e.g. ownership). During the course of the LDRD, we focused on Electric Vehicle charging stations and this information is available via the Department of Energy (DOE) database on fueling stations maintained by NREL. Within the context of the DOE CESER project, we have focused on Battery Energy Storage Systems (BESS). Second, the Network Extraction component will iteratively construct a social network graph given the set of organizations and people extracted in the previous step. Organizations (and eventually People if desired) are then fed as a query to the `topgear-construct_social_network` Apache Airflow workflow which given a set of initial companies and data sets (e.g. SEC EDGAR form types, OpenCorporates, Crunchbase). This Airflow workflow will iteratively query such data sources to discover relationships with new organizations and people. For example, this module can iteratively query SEC EDGAR for metadata that documents the number of each type of form for the given set of companies and their location. This forms metadata represents a catalog of data sources from SEC EDGAR for the extracted social network knowledge graph. The pipeline then downloads these forms from the website and saves them in a build directory for further processing. These documents are then parsed for entities and relations. Again, we note that in additional to SEC data sources, this step can also pull in information on organizations via API services such as CrunchBase and OpenCorporates or bulk data sources. At the end of this step, the resultant social network, the Critical Infrastructure network, and the edges that encode relationships between organizations and CI facilities, form the Adversarial Socio-Technical Network (ASTN) that informs the analysis. Third, the Analysis component processes these generated ASTN. Previously, that has included the ability to compare prevalence of different vendors for a given infrastructure component type across different regions as well as identify common public and private investors across those vendors. This was demonstrated for EV Charging Stations across several different metropolitan areas within an IEEE PES GridEdge publication. More recently, we have looked at ways to identify infrastructure owners and operators of BESS with the most nameplate capacity across different states as well as other indictors of risk resulting from changes in ownership over time. Finally, the Visualization component consists of an HTML/CSS/JS framework by which users can interact geospatial, operational, and organizational relationships across a given portfolio of Critical Infrastructure facilities. The objective is to provide a library of UI/UX modules that can be repurposed for stakeholder-specific dashboards. All of the modules are related via a common event model that enables UI actions in one view to percolate across the other views.

Weaver, Gabriel [Idaho National Laboratory (INL), ↗

Stacked filters: learning to filter by structure

We present Stacked Filters, a new probabilistic filter which is fast and robust similar to query-agnostic filters (such as Bloom and Cuckoo filters), and at the same time brings low false positive rates and sizes similar to classifier-based filters (such as Learned Filters). The core idea is that Stacked Filters incorporate workload knowledge about frequently queried non-existing values. Instead of learning, they structurally incorporate that knowledge using hashing and several sequenced filter layers, indexing both data and frequent negatives. Stacked Filters can also gather workload knowledge on-the-fly and adaptively build the filter. We show experimentally that for a given memory budget, Stacked Filters achieve end-to-end query throughput up to 130x better than the best alternative for a workload, either query-agnostic or classifier-based filters, and depending on where data is (SSD or HDD).

Computer Science↗

Leveraging Application Data Constraints to Optimize Database-Backed Web Applications

Exploiting the relationships among data is a classical query optimization technique. As persistent data is increasingly being created and maintained programmatically, prior work that infers data relationships from data statistics misses an important opportunity. We present Coco, the first tool that identifies data relationships by analyzing database-backed applications. Once identified, Coco leverages the constraints to optimize the application's physical design and query execution. Instead of developing a fixed set of predefined rewriting rules, Coco employs an enumerate-test-verify technique to automatically exploit the discovered data constraints to improve query execution. Each resulting rewrite is provably equivalent to the original query. Using 14 real-world web applications, our experiments show that Coco can discover numerous data constraints from code analysis and improve real-world application performance significantly.

Computer Science↗

WELLS Interactive Application

The Wellbore Exploration and Location Logistic System (WELLS) Interactive Application is an interactive tool to enable easy exploration and visualization of the living national wellbore database (WELLS Database (https://edx.netl.doe.gov/dataset/wells_database)). The tool and underlying database were created and are maintained by the National Energy Technology Laboratory (NETL), providing visualization of the more than six million public wellbore records from more than 65 authoritative state, federal, and tribal resources. The WELLS Interactive Application serves up wellbore data from oil, gas, underground injection, research, geothermal, geotechnical, groundwater, and other types of wells in a single, standardized, unified system. In addition to the surface location of these wells, the underlying database combines select key attributes for features such as well age, depth, and operating status. The system also provides users with references back to the original sources used in this unified platform. The underlying data can be accessed through the WELLS Database: https://edx.netl.doe.gov/dataset/wells_database Additional Information: The WELLS Interactive Application (formerly titled CO2-Locate) enables visualization and access to the public wellbore records through an intuitive web-based mapping tool. The WELLS Interactive Application was designed to help users visualize, query, analyze, and download wellbore records. Public wellbore points are included as a layer in the Map page, called Public Wells. Additionally, a multivariate hexagon grid summarizing well density from proprietary well data, called Well Density, is included to identify data gaps between the public and proprietary well data. Filtering functionalities in the tool allow these two layers to be spatially filtered by state, county, or basin as well as by status, type, true vertical depth, and spud year. The WELLS Interactive Application also contains a Near Me tool can be used to search and explore wellbore data within a user-defined distance of a specified location on the map, which can also be downloaded. The Query tool allows users to query the selected or filtered wells in the Public Wells layer and export the data. For additional information on these tool functionalities, see the help documentation on the About page of the tool. Notes for Consideration: The Well Density layer provided in this application is derived from proprietary wellbore data, the records of which do not always contain values for key features (status, type, true vertical depth, or spud year). Therefore, data might not be available when layers are queried for all filter combinations. Additionally, visualizing layers and applying filters may take additional time to load (i.e., draw on the map) due to the large size of the data.

ccs↗

Adding Data Management Services to Parallel File Systems

The objective of this project, called DAMASC for “Data Management in Scientific Computing”, is to coalesce data management with parallel file system management to present a declarative interface to scientists for managing, querying, and analyzing extremely large data sets efficiently and predictably. Managing extremely large data sets is a key challenge of exascale computing. The overhead, energy, and cost of moving massive volumes of data demand designs where computation is close to storage. In current architectures, compute/analysis clusters access data in a physically separate parallel file system and largely leave it scientist to reduce data movement. Over the past decades the high-end computing community has adopted middleware with multiple layers of abstractions and specialized file formats such as NetCDF-4 and HDF5. These abstractions provide a limited set of high-level data processing functions, but have inherent functionality and performance limitations: middleware that provides access to the highly structured contents of scientific data files stored in the (unstructured) file systems can only optimize to the extent that file system interfaces permit; the highly structured formats of these files often impedes native file system performance optimizations. We are developing Damasc, an enhanced high-performance file system with native rich data management services. Damasc will enable efficient queries and updates over files stored in their native byte-stream format while retaining the inherent performance of file system data storage via declarative queries and updates over views of underlying files. Damasc has four key benefits for the development of data-intensive scientific code: (1) applications can use important data-management services, such as declarative queries, views, and provenance tracking, that are currently available only within database systems; (2) the use of these services becomes easier, as they are provided within a familiar file-based ecosystem; (3) common optimizations, e.g., indexing and caching, are readily supported across several file formats, avoiding effort duplication; and (4) performance improves significantly, as data processing is integrated more tightly with data storage. Our key contributions are: SciHadoop which explores changes to MapReduce assumption by taking advantage of semantics of structured data while preserving MapReduce’s failure and resource management; DataMods which extends common abstractions of parallel file systems so they become programmable such that they can be extended to natively support a variety of data models and can be hooked into emerging distributed runtimes such as Stanford’s Legion; and Miso which combines Hadoop and relational data warehousing to minimize time to insight, taking into account the overhead of ingesting data into data warehousing.

96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗

A Surveillance Infrastructure for Malaria Analytics: Provisioning Data Access and Preservation of Interoperability

Background: According to the World Health Organization, malaria surveillance is weakest in countries and regions with the highest malaria burden. A core obstacle is that the data required to perform malaria surveillance are fragmented in multiple data silos distributed across geographic regions. Furthermore, consistent integrated malaria data sources are few, and a low degree of interoperability exists between them. As a result, it is difficult to identify disease trends and to plan for effective interventions. Objective: We propose the Semantics, Interoperability, and Evolution for Malaria Analytics (SIEMA) platform for use in malaria surveillance based on semantic data federation. Using this approach, it is possible to access distributed data, extend and preserve interoperability between multiple dynamic distributed malaria sources, and facilitate detection of system changes that can interrupt mission-critical global surveillance activities. Methods: We used Semantic Automated Discovery and Integration (SADI) Semantic Web Services to enable data access and improve interoperability, and the graphical user interface-enabled semantic query engine HYDRA to implement the target queries typical of malaria programs. We implemented a custom algorithm to detect changes to community-developed terminologies, data sources, and services that are core to SIEMA. This algorithm reports to a dashboard. Valet SADI is used to mitigate the impact of changes by rebuilding affected services. Results: We developed a prototype surveillance and change management platform from a combination of third-party tools, community-developed terminologies, and custom algorithms. We illustrated a methodology and core infrastructure to facilitate interoperable access to distributed data sources using SADI Semantic Web services. This degree of access makes it possible to implement complex queries needed by our user community with minimal technical skill. We implemented a dashboard that reports on terminology changes that can render the services inactive, jeopardizing system interoperability. Using this information, end users can control and reactively rebuild services to preserve interoperability and minimize service downtime. Conclusions: We introduce a framework suitable for use in malaria surveillance that supports the creation of flexible surveillance queries across distributed data resources. The platform provides interoperable access to target data sources, is domain agnostic, and with updates to core terminological resources is readily transferable to other surveillance activities. A dashboard enables users to review changes to the infrastructure and invoke system updates. The platform significantly extends the range of functionalities offered by malaria information systems, beyond the state-of-the-art.

60 APPLIED LIFE SCIENCES↗

A Selectivity based approach to Continuous Pattern Detection in Streaming Graphs

Cyber security is one of the most significant technical challenges in current times. Detecting adversarial activities, prevention of theft of intellectual properties and customer data is a high priority for corporations and government agencies around the world. Cyber defenders need to analyze massive-scale, high-resolution network flows to identify, categorize, and mitigate attacks involving net- works spanning institutional and national boundaries. Many of the cyber attacks can be described as subgraph patterns, with promi- nent examples being insider infiltrations (path queries), denial of service (parallel paths) and malicious spreads (tree queries). This motivates us to explore subgraph matching on streaming graphs in a continuous setting. The novelty of our work lies in using the subgraph distributional statistics collected from the streaming graph to determine the query processing strategy. We introduce a “Lazy Search" algorithm where the search strategy is decided on a vertex-to-vertex basis depending on the likelihood of a match in the vertex neighborhood. We also propose a metric named “Relative Selectivity" that is used to se- lect between different query processing strategies. Our experiments performed on real online news, network traffic stream and a syn- thetic social network benchmark demonstrate 10-100x speedups over selectivity agnostic approaches.

graphs, cybersecurity↗

Using a data base management system for modelling SSME test history data

The usefulness of a data base management system (DBMS) for modelling historical test data for the complete series of static test firings for the Space Shuttle Main Engine (SSME) was assessed. From an analysis of user data base query requirements, it became clear that a relational DMBS which included a relationally complete query language would permit a model satisfying the query requirements. Representative models and sample queries are discussed. A list of environment-particular evaluation criteria for the desired DBMS was constructed; these criteria include requirements in the areas of user-interface complexity, program independence, flexibility, modifiability, and output capability. The evaluation process included the construction of several prototype data bases for user assessement. The systems studied, representing the three major DBMS conceptual models, were: MIRADS, a hierarchical system; DMS-1100, a CODASYL-based network system; ORACLE, a relational system; and DATATRIEVE, a relational-type system.

Abernethy, K.↗

Semantic based man-machine interface for real-time communication

A flight expert system (FLES) was developed to assist pilots in monitoring, diagnosing and recovering from in-flight faults. To provide a communications interface between the flight crew and FLES, a natural language interface (NALI) was implemented. Input to NALI is processed by three processors: (1) the semantics parser; (2) the knowledge retriever; and (3) the response generator. First the semantic parser extracts meaningful words and phrases to generate an internal representation of the query. At this point, the semantic parser has the ability to map different input forms related to the same concept into the same internal representation. Then the knowledge retriever analyzes and stores the context of the query to aid in resolving ellipses and pronoun references. At the end of this process, a sequence of retrievel functions is created as a first step in generating the proper response. Finally, the response generator generates the natural language response to the query. The architecture of NALI was designed to process both temporal and nontemporal queries. The architecture and implementation of NALI are described.

Ali, M.↗

Database Reports Over the Internet

Most of the summer was spent developing software that would permit existing test report forms to be printed over the web on a printer that is supported by Adobe Acrobat Reader. The data is stored in a DBMS (Data Base Management System). The client asks for the information from the database using an HTML (Hyper Text Markup Language) form in a web browser. JavaScript is used with the forms to assist the user and verify the integrity of the entered data. Queries to a database are made in SQL (Sequential Query Language), a widely supported standard for making queries to databases. Java servlets, programs written in the Java programming language running under the control of network server software, interrogate the database and complete a PDF form template kept in a file. The completed report is sent to the browser requesting the report. Some errors are sent to the browser in an HTML web page, others are reported to the server. Access to the databases was restricted since the data are being transported to new DBMS software that will run on new hardware. However, the SQL queries were made to Microsoft Access, a DBMS that is available on most PCs (Personal Computers). Access does support the SQL commands that were used, and a database was created with Access that contained typical data for the report forms.

Dean Lance Smith↗

Developing A Web-based User Interface for Semantic Information Retrieval

While there are now a number of languages and frameworks that enable computer-based systems to search stored data semantically, the optimal design for effective user interfaces for such systems is still uncle ar. Such interfaces should mask unnecessary query detail from users, yet still allow them to build queries of arbitrary complexity without significant restrictions. We developed a user interface supporting s emantic query generation for Semanticorganizer, a tool used by scient ists and engineers at NASA to construct networks of knowledge and dat a. Through this interface users can select node types, node attribute s and node links to build ad-hoc semantic queries for searching the S emanticOrganizer network.

Berrios, Daniel C.↗

Robust Requirements Tracing via Internet Search Technology: Improving an IV and V Technique

There are three major objectives to this phase of the work. (1) Improvement of Information Retrieval (IR) methods for Independent Verification and Validation (IV&V) requirements tracing. Information Retrieval methods are typically developed for very large (order of millions - tens of millions and more documents) document collections and therefore, most successfully used methods somewhat sacrifice precision and recall in order to achieve efficiency. At the same time typical IR systems treat all user queries as independent of each other and assume that relevance of documents to queries is subjective for each user. The IV&V requirements tracing problem has a much smaller data set to operate on, even for large software development projects; the set of queries is predetermined by the high-level specification document and individual requirements considered as query input to IR methods are not necessarily independent from each other. Namely, knowledge about the links for one requirement may be helpful in determining the links of another requirement. Finally, while the final decision on the exact form of the traceability matrix still belongs to the IV&V analyst, his/her decisions are much less arbitrary than those of an Internet search engine user. All this suggests that the information available to us in the framework of the IV&V tracing problem can be successfully leveraged to enhance standard IR techniques, which in turn would lead to increased recall and precision. We developed several new methods during Phase II; (2) IV&V requirements tracing IR toolkit. Based on the methods developed in Phase I and their improvements developed in Phase II, we built a toolkit of IR methods for IV&V requirements tracing. The toolkit has been integrated, at the data level, with SAIC's SuperTracePlus (STP) tool; (3) Toolkit testing. We tested the methods included in the IV&V requirements tracing IR toolkit on a number of projects.

Hayes, Jane↗

EXTENSIBLE DATABASE FRAMEWORK FOR MANAGEMENT OF UNSTRUCTURED AND SEMI-STRUCTURED DOCUMENTS

Method and system for querying a collection of Unstructured or semi-structured documents to identify presence of, and provide context and/or content for, keywords and/or keyphrases. The documents are analyzed and assigned a node structure, including an ordered sequence of mutually exclusive node segments or strings. Each node has an associated set of at least four, five or six attributes with node information and can represent a format marker or text, with the last node in any node segment usually being a text node. A keyword (or keyphrase) is specified. and the last node in each node segment is searched for a match with the keyword. When a match is found at a query node, or at a node determined with reference to a query node, the system displays the context andor the content of the query node.

Gawdiak, Yuri O.↗

Efficient Implementation of an Optimal Interpolator for Large Spatial Data Sets

Scattered data interpolation is a problem of interest in numerous areas such as electronic imaging, smooth surface modeling, and computational geometry. Our motivation arises from applications in geology and mining, which often involve large scattered data sets and a demand for high accuracy. The method of choice is ordinary kriging. This is because it is a best unbiased estimator. Unfortunately, this interpolant is computationally very expensive to compute exactly. For n scattered data points, computing the value of a single interpolant involves solving a dense linear system of size roughly n x n. This is infeasible for large n. In practice, kriging is solved approximately by local approaches that are based on considering only a relatively small'number of points that lie close to the query point. There are many problems with this local approach, however. The first is that determining the proper neighborhood size is tricky, and is usually solved by ad hoc methods such as selecting a fixed number of nearest neighbors or all the points lying within a fixed radius. Such fixed neighborhood sizes may not work well for all query points, depending on local density of the point distribution. Local methods also suffer from the problem that the resulting interpolant is not continuous. Meyer showed that while kriging produces smooth continues surfaces, it has zero order continuity along its borders. Thus, at interface boundaries where the neighborhood changes, the interpolant behaves discontinuously. Therefore, it is important to consider and solve the global system for each interpolant. However, solving such large dense systems for each query point is impractical. Recently a more principled approach to approximating kriging has been proposed based on a technique called covariance tapering. The problems arise from the fact that the covariance functions that are used in kriging have global support. Our implementations combine, utilize, and enhance a number of different approaches that have been introduced in literature for solving large linear systems for interpolation of scattered data points. For very large systems, exact methods such as Gaussian elimination are impractical since they require 0(n(exp 3)) time and 0(n(exp 2)) storage. As Billings et al. suggested, we use an iterative approach. In particular, we use the SYMMLQ method, for solving the large but sparse ordinary kriging systems that result from tapering. The main technical issue that need to be overcome in our algorithmic solution is that the points' covariance matrix for kriging should be symmetric positive definite. The goal of tapering is to obtain a sparse approximate representation of the covariance matrix while maintaining its positive definiteness. Furrer et al. used tapering to obtain a sparse linear system of the form Ax = b, where A is the tapered symmetric positive definite covariance matrix. Thus, Cholesky factorization could be used to solve their linear systems. They implemented an efficient sparse Cholesky decomposition method. They also showed if these tapers are used for a limited class of covariance models, the solution of the system converges to the solution of the original system. Matrix A in the ordinary kriging system, while symmetric, is not positive definite. Thus, their approach is not applicable to the ordinary kriging system. Therefore, we use tapering only to obtain a sparse linear system. Then, we use SYMMLQ to solve the ordinary kriging system. We show that solving large kriging systems becomes practical via tapering and iterative methods, and results in lower estimation errors compared to traditional local approaches, and significant memory savings compared to the original global system. We also developed a more efficient variant of the sparse SYMMLQ method for large ordinary kriging systems. This approach adaptively finds the correct local neighborhood for each query point in the interpolation process.

Memarsadeghi, Nargess↗

Active Learning with Irrelevant Examples

An improved active learning method has been devised for training data classifiers. One example of a data classifier is the algorithm used by the United States Postal Service since the 1960s to recognize scans of handwritten digits for processing zip codes. Active learning algorithms enable rapid training with minimal investment of time on the part of human experts to provide training examples consisting of correctly classified (labeled) input data. They function by identifying which examples would be most profitable for a human expert to label. The goal is to maximize classifier accuracy while minimizing the number of examples the expert must label. Although there are several well-established methods for active learning, they may not operate well when irrelevant examples are present in the data set. That is, they may select an item for labeling that the expert simply cannot assign to any of the valid classes. In the context of classifying handwritten digits, the irrelevant items may include stray marks, smudges, and mis-scans. Querying the expert about these items results in wasted time or erroneous labels, if the expert is forced to assign the item to one of the valid classes. In contrast, the new algorithm provides a specific mechanism for avoiding querying the irrelevant items. This algorithm has two components: an active learner (which could be a conventional active learning algorithm) and a relevance classifier. The combination of these components yields a method, denoted Relevance Bias, that enables the active learner to avoid querying irrelevant data so as to increase its learning rate and efficiency when irrelevant items are present. The algorithm collects irrelevant data in a set of rejected examples, then trains the relevance classifier to distinguish between labeled (relevant) training examples and the rejected ones. The active learner combines its ranking of the items with the probability that they are relevant to yield a final decision about which item to present to the expert for labeling. Experiments on several data sets have demonstrated that the Relevance Bias approach significantly decreases the number of irrelevant items queried and also accelerates learning speed.

Wagstaff, Kiri↗