Abstract: A data analytics platform is provided for forecasting future states of commodities and other assets, based on processing of both textual and numerical data sources. The platform includes a multi-layer machine learning-based model that extracts sentiment from textual data in a natural language processing engine, evaluates numerical data in a time-series analysis, and generates an initial forecast for the commodity or asset being analyzed. The platform includes multiple applications of neural networks to develop augmented forecasts from further analysis of relevant information as it is collected. These include commodity-specific neural networks designed to continually develop taxonomies used to process commodity sentiment, and applications of reinforcement learning, symbolic networks, and unsupervised meta learning to improve overall performance and accuracy of the forecasts generated.

Abstract: A data analytics platform is provided for forecasting future states of commodities and other assets, based on processing of both textual and numerical data sources. The platform includes a multi-layer machine learning-based model that extracts sentiment from textual data in a natural language processing engine, evaluates numerical data in a time-series analysis, and generates an initial forecast for the commodity or asset being analyzed. The platform includes multiple applications of neural networks to develop augmented forecasts from further analysis of relevant information as it is collected. These include commodity-specific neural networks designed to continually develop taxonomies used to process commodity sentiment, and applications of reinforcement learning, symbolic networks, and unsupervised meta learning to improve overall performance and accuracy of the forecasts generated.

Abstract: A data analytics platform is provided for forecasting future states of commodities and other assets, based on processing of both textual and numerical data sources. The platform includes a multi-layer machine learning-based model that extracts sentiment from textual data in a natural language processing engine, evaluates numerical data in a time-series analysis, and generates an initial forecast for the commodity or asset being analyzed. The platform includes multiple applications of neural networks to develop augmented forecasts from further analysis of relevant information as it is collected. These include commodity-specific neural networks designed to continually develop taxonomies used to process commodity sentiment, and applications of reinforcement learning, symbolic networks, and unsupervised meta learning to improve overall performance and accuracy of the forecasts generated.

Abstract: A data analytics platform is provided for forecasting future states of commodities and other assets, based on processing of both textual and numerical data sources. The platform includes a multi-layer machine learning-based model that extracts sentiment from textual data in a natural language processing engine, evaluates numerical data in a time-series analysis, and generates an initial forecast for the commodity or asset being analyzed. The platform includes multiple applications of neural networks to develop augmented forecasts from further analysis of relevant information as it is collected. These include commodity-specific neural networks designed to continually develop taxonomies used to process commodity sentiment, and applications of reinforcement learning, symbolic networks, and unsupervised meta learning to improve overall performance and accuracy of the forecasts generated.

Abstract: Detection and identification a field's boundaries is performed in a workflow based on processing images of the field captured at different times, relative to a defined seed point. Images are clipped to align with the seed point and a bounding box around the seed point, and a mask is built by extracting edges of the field from the images. The workflow floods an area around the seed point that has pixels of a similar color, using the mask as an initial boundary. The flooded area is compared to threshold parameter values, which are tuned to refine the identified boundary. Flooded areas in multiple images are combined, and a boundary is built based on the combined flooded set. Manual, interactive tuning of floodfill areas allows for a separate boundary detection and identification workflow or for refinement of the automatic boundary detection workflow.

Abstract: Detection and identification a field's boundaries is performed in a workflow based on processing images of the field captured at different times, relative to a defined seed point. Images are clipped to align with the seed point and a bounding box around the seed point, and a mask is built by extracting edges of the field from the images. The workflow floods an area around the seed point that has pixels of a similar color, using the mask as an initial boundary. The flooded area is compared to threshold parameter values, which are tuned to refine the identified boundary. Flooded areas in multiple images are combined, and a boundary is built based on the combined flooded set. Manual, interactive tuning of floodfill areas allows for a separate boundary detection and identification workflow or for refinement of the automatic boundary detection workflow.

Abstract: Detection and identification a field's boundaries is performed in a workflow based on processing images of the field captured at different times, relative to a defined seed point. Images are clipped to align with the seed point and a bounding box around the seed point, and a mask is built by extracting edges of the field from the images. The workflow floods an area around the seed point that has pixels of a similar color, using the mask as an initial boundary. The flooded area is compared to threshold parameter values, which are tuned to refine the identified boundary. Flooded areas in multiple images are combined, and a boundary is built based on the combined flooded set. Manual, interactive tuning of floodfill areas allows for a separate boundary detection and identification workflow or for refinement of the automatic boundary detection workflow.

Abstract: Detection and identification a field's boundaries is performed in a workflow based on processing images of the field captured at different times, relative to a defined seed point. Images are clipped to align with the seed point and a bounding box around the seed point, and a mask is built by extracting edges of the field from the images. The workflow floods an area around the seed point that has pixels of a similar color, using the mask as an initial boundary. The flooded area is compared to threshold parameter values, which are tuned to refine the identified boundary. Flooded areas in multiple images are combined, and a boundary is built based on the combined flooded set. Manual, interactive tuning of floodfill areas allows for a separate boundary detection and identification workflow or for refinement of the automatic boundary detection workflow.

Abstract: Quality assessment of probe data collected from GPS systems is performed by a system and method of determining a value of data points provided by different vendors of such data. Incoming raw probe data is initially analyzed for removal of extraneous data points, and is then mapped to roadway links and smoothed out. The resulting output is processed to determine the coverage value of data provided by a given vendor and enable a comparison between different vendors. Such a model of probe data processing also enables an evaluation of a contribution of further vendors of raw probe data to an existing dataset. Additionally, a real-time performance evaluation of continually-ingested probe data includes building historical and data count profiles, and generating output data represented by a number of data points for a specific distance within a geo-box representing a geographical area, to project a value of raw probe data for a next incremental time period.

Abstract: Estimation of traffic speed includes applying data processing functions to determine missing speed information by smoothing spatial and temporal GPS data to achieve an accurate estimation of link speed over all links of a transportation network at all time periods. This estimation of traffic speed uses one link's observed speed information to estimate neighboring links without observed speed information and therefore provides a system and method of processing collected GPS data to obtain a thorough understanding of traffic flow conditions for all represented links without further collection of GPS data. The present invention also provides a framework for analyzing and improving real-time collection of GPS speed data.

Abstract: A framework for performance evaluation and active management of a transportation network infrastructure reconstructs traffic flow profiles by modeling annual average daily traffic data and collected traffic speed data to estimate an hourly traffic flow profile for a roadway segment, or link. Total daily flow for a link is derived from the corresponding annual average daily traffic data for that link, and is adjusting by the day of week and the monthly seasonal factors. An hourly flow distribution profile for a roadway link is then constructed using the traffic speed data relative to that link.

Abstract: A framework for performance evaluation and active management of a transportation network infrastructure reconstructs traffic flow profiles by modeling annual average daily traffic data and collected traffic speed data to estimate an hourly traffic flow profile for a roadway segment, or link. Total daily flow for a link is derived from the corresponding annual average daily traffic data for that link, and is adjusting by the day of week and the monthly seasonal factors. An hourly flow distribution profile for a roadway link is then constructed using the traffic speed data relative to that link.

Abstract: Estimation of traffic speed includes applying data processing functions to determine missing speed information by smoothing spatial and temporal GPS data to achieve an accurate estimation of link speed over all links of a transportation network at all time periods. This estimation of traffic speed uses one link's observed speed information to estimate neighboring links without observed speed information and therefore provides a system and method of processing collected GPS data to obtain a thorough understanding of traffic flow conditions for all represented links without further collection of GPS data. The present invention also provides a framework for analyzing and improving real-time collection of GPS speed data.

Abstract: Quality assessment of probe data collected from GPS systems is performed by a system and method of determining a value of data points provided by different vendors of such data. Incoming raw probe data is initially analyzed for removal of extraneous data points, and is then mapped to roadway links and smoothed out. The resulting output is processed to determine the coverage value of data provided by a given vendor and enable a comparison between different vendors. Such a model of probe data processing also enables an evaluation of a contribution of further vendors of raw probe data to an existing dataset. Additionally, a real-time performance evaluation of continually-ingested probe data includes building historical and data count profiles, and generating output data represented by a number of data points for a specific distance within a geo-box representing a geographical area, to project a value of raw probe data for a next incremental time period.