Shrutilipi Bhattacharjee

Spatial Interpolation to Predict Missing Attributes in GIS Using Semantic Kriging

Prediction of spatial attributes has attracted significant research interest in recent years. It is challenging especially when spatial data contain errors and missing values. Geostatistical estimators are used to predict the missing attribute values from the observed values of known surrounding data points, a general form of which is referred as kriging in the field of geographic information system and remote sensing. The proposed semantic kriging ( SemK) tries to blend the semantics of spatial features (of surrounding data points) with ordinary kriging (OK) method for prediction of the attribute. Experimentation has been carried out with land surface temperature data of four major metropolitan cities in India. It shows that SemK outperforms the OK and most of the existing spatial interpolation methods.

A Spatial Web Crawler for Discovering Geo-servers and Semantic Referencing with Spatial Features

Improvement of technologies in the field of spatial data collection provide a lot of research opportunities in the field of Geographic Information System. The geospatial data are often dynamic in nature and available in heterogeneous format. The online spatial data sources are one of the key avenues for publishing and retrieving the geo-spatial data. Efficient discovery of these data sources through Internet, retrieval and analysis of useful information, is one of major challenges in this field. The paper proposes a framework for discovering the geo-spatial data sources using a spatial web crawler. This will facilitate processing of spatial queries involving distributed heterogeneous data repositories. This is being done with the help of the Web Feature Service WFS standard specification provided by Open Geospatial Consortium OGC. Geo-spatial information is retrieved further for semantic annotations of the data sources using ontology. The semantic information is stored in the form of feature_type repositories as the area of interest lies around the geographic features provided by the geo-servers. The performance study analyzes the accuracy of discovery and semantic annotation of geo-servers for better understanding the framework.

Ontology based framework for semantic resolution of geospatial query

Increasing availability of geospatial data provides exceptional opportunities in knowledge creation and distribution. For the discovery of suitable data sources, keyword based search in the catalogue becomes inaccurate. The main reason behind this is the existing semantic heterogeneity in the database schema, deployed by different service providers. It necessitates the semantic management of the spatial catalogues. This paper presents an ontology based approach which is useful to create and manage catalogues semantically, hence resolving the semantic heterogeneity between geospatial repositories and incompatibility with spatial queries. Further, there is no standard available for semantic searching of the spatial catalogue till date. It will enhance the data extraction process by providing semantic meaning to the spatial catalogue.

Performance Evaluation of Semantic Kriging: A Euclidean Vector Analysis Approach

Prediction of spatial attributes in geospatial data repositories is indispensable in the field of remote sensing and geographic information system. The semantic kriging (SemK) approach semantically captures the domain knowledge of the terrain in terms of local spatial features for spatial attribute prediction. It produces better results than ordinary kriging and other prediction methods. This letter focuses on the theoretical and empirical analyses of the SemK. A Euclidean vector analysis approach is adopted to theoretically prove the efficacy of SemK in capturing semantic knowledge.

Ontology based spatial clustering framework for implicit knowledge discovery

Increasing availability of geo-spatial data, enabled by improvement of technologies in the field of spatial data collection, provides unprecedented opportunities in knowledge creation and distribution. Beside the integration of distributed spatial data, available in Spatial Data Infrastructures (SDIs), proper processing of data is crucial for knowledge discovery. This work proposes an ontology based spatial knowledge discovery framework based on Open Geospatial Consortium (OGC) standards. Data processing in geo-spatial domain is done using spatial clustering algorithm. It is implemented as the web processing service of this framework. Prediction of urban house price is taken as the case study.

Automatic Resolution of Semantic Heterogeneity in GIS: An Ontology Based Approach

To facilitate the access of geographic information, spatial data infrastructures (SDI) are being set up within regions, countries and internationally. Consequently, the interoperable access of the information across disparate geospatial databases has become essential. Semantic heterogeneity is one of the crucial issues to get resolved for efficient and accurate retrieval of spatial data. This work has adopted an ontology based approach to overcome the semantic heterogeneities, usually exists in different heterogeneous data repositories. The proposed method is used for semantic matching between user query concepts and the target concepts in the spatial databases. It overcomes the problem of manual intervention during matching process; hence making it automated.

Time-series augmentation of semantic kriging for the prediction of meteorological parameters

Spatio-temporal pattern analysis of meteorological parameters has been studied extensively in the field of remote sensing (RS) and geographic information system (GIS). It is an important data mining strategy for modeling the temporal dynamics of these parameters and forecasting them in future time instances. The meteorological parameters, measured near the earth surface, are eminently influenced by the surrounding land-cover distribution of the terrain. For the time-series prediction of these parameters, the knowledge of spatial land-covers can be contemplated within the space-time trade-off between the sample points. This work presents a new kriging based spatio-temporal interpolation method, namely times-series semantic kriging (SemKts) which deals with land-cover dynamics and incorporates this knowledge for better prediction accuracy. It is a time-series extension of our earlier work on semantic kriging (SemK) for spatial interpolation [1] [2]. Experimentation has been carried out by considering real land surface temperature data in the spatial region Kolkata, India. It has been observed that the semantically enhanced space-time trade-off analysis by SemKts yields more accurate result than most of the popular methods for prediction.

Analysis of Spatial Autocorrelation for Traffic Accident Data Based on Spatial Decision Tree

With rapid increase of scope, coverage and volume of geographic datasets, knowledge discovery from spatial data have drawn a lot of research interest for last few decades. Traditional analytical techniques cannot easily discover new, implicit patterns, and relationships that are hidden into geographic datasets. The principle of this work is to evaluate the performance of traditional and spatial data mining techniques for analysing spatial certainty, such as spatial autocorrelation. Analysis is done by classification technique, i.e. a Decision Tree (DT) based approach on a spatial diversity coefficient. ID3 (Iterative Dichotomiser 3) algorithm is used for building the conventional and spatial decision trees. A synthetically generated spatial accident dataset and real accident dataset are used for this purpose. The spatial DT (SDT) is found to be more significant in spatial decision making.

Measurement of Semantic Similarity: A Concept Hierarchy Based Approach

Resolving semantic heterogeneity is one of the major issues in many fields, namely, natural language processing, search engine development, document clustering, geospatial information retrieval and knowledge discovery, etc. Semantic heterogeneity is often considered as an obstacle for realizing full interoperability among diverse datasets. Appropriate measurement metric is essential to properly understand the extent of similarity between concepts. The proposed approach is based on the notion of concept hierarchy which is built using a lexical database. The WordNet, a semantic lexical database, is used here to build the semantic hierarchy. A measurement metric is also proposed to quantify the extent of similarity between a pair of concepts. The work is compared with existing methodologies on Miller-Charles benchmark dataset using three correlation coefficients (Pearson’s, Spearman’s and Kendall Tau rank correlation coefficients). The proposed approach is found to yield better results than most of the existing techniques.

Prediction of meteorological parameters: an a-posteriori probabilistic semantic kriging approach

Meteorological parameters are often considered as crucial factors for climatological pattern analysis. Predictions of these parameters have been studied extensively in the field of remote sensing and GIS. It is one of the most critical steps involved in most of the meteorological data mining process. Spatial interpolation is an efficient technique to yield minimal error in prediction. From existing literatures, it is evident that the land-use/land-cover (LULC) distribution of the terrain influences these parameters in a varying manner and it is important to model their behaviour for climatological analyses. However, this semantic LULC knowledge of the terrain is generally ignored in the prediction process of the meteorological parameters. Recently, we have proposed a new spatial interpolation technique, namely semantic kriging (SemK) [3,5,7], which considers the semantic LULC knowledge for land-atmospheric interaction modeling and incorporates it into the existing interpolation process for better accuracy. However, the a-priori correlation analysis of SemK ignores the effect of other nearby LULC classes on each other. This article presents a new variant of SemK, namely a-posterior probabilistic Bayesian SemK (BSemK), which extends the a-priori correlation analysis of SemK with a-posterior probabilistic analysis. The proposed approach provides more accurate estimation of the parameters. Experimentation with LST data advocates the efficacy of the proposed approach compared to the a-priori SemK and other existing interpolation techniques.