Limits...
Harnessing ontology and machine learning for RSO classification

View Article: PubMed Central - PubMed

ABSTRACT

Classification is an important part of resident space objects (RSOs) identification, which is a main focus of space situational awareness. Owing to the absence of some features caused by the limited and uncertain observations, RSO classification remains a difficult task. In this paper, an ontology for RSO classification named OntoStar is built upon domain knowledge and machine learning rules. Then data describing RSO are represented by OntoStar. A demo shows how an RSO is classified based on OntoStar. It is also shown in the demo that traceable and comprehensible reasons for the classification can be given, hence the classification can be checked and validated. Experiments on WEKA show that ontology-based classification gains a relatively high accuracy and precision for classifying RSOs. When classifying RSOs with imperfect data, ontology-based classification keeps its performances, showing evident advantages over classical machine learning classifiers who either have increases of 5 % at least in FP rate or have decreases of 5 % at least in indexes such as accuracy, precision and recall.

No MeSH data available.


Justification parsed to tree-like structure
© Copyright Policy - OpenAccess
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC5037103&req=5

Fig5: Justification parsed to tree-like structure

Mentions: In Fig. 4, the imported data which describe inst2cls are shown in the middle box and lower box of the left part. The given knowledge about inst2cls contains information of owner, orbit, rcs, height and power of inst2cls, and the statement that inst2cls is an RSO. The statement that inst2cls is a Reconnaissance_Satellite, shown in the dot-line box of the lower left part, is derived from the given knowledge through ontology reasoning. A classification justification is found by the service explanation, and the classification justification is shown in the right part. With the classification justification being comprehensible, the deriving process from pre-condition to conclusion can be parsed into a intelligible tree-like structure, as shown in Fig. 5.Fig. 5


Harnessing ontology and machine learning for RSO classification
Justification parsed to tree-like structure
© Copyright Policy - OpenAccess
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC5037103&req=5

Fig5: Justification parsed to tree-like structure
Mentions: In Fig. 4, the imported data which describe inst2cls are shown in the middle box and lower box of the left part. The given knowledge about inst2cls contains information of owner, orbit, rcs, height and power of inst2cls, and the statement that inst2cls is an RSO. The statement that inst2cls is a Reconnaissance_Satellite, shown in the dot-line box of the lower left part, is derived from the given knowledge through ontology reasoning. A classification justification is found by the service explanation, and the classification justification is shown in the right part. With the classification justification being comprehensible, the deriving process from pre-condition to conclusion can be parsed into a intelligible tree-like structure, as shown in Fig. 5.Fig. 5

View Article: PubMed Central - PubMed

ABSTRACT

Classification is an important part of resident space objects (RSOs) identification, which is a main focus of space situational awareness. Owing to the absence of some features caused by the limited and uncertain observations, RSO classification remains a difficult task. In this paper, an ontology for RSO classification named OntoStar is built upon domain knowledge and machine learning rules. Then data describing RSO are represented by OntoStar. A demo shows how an RSO is classified based on OntoStar. It is also shown in the demo that traceable and comprehensible reasons for the classification can be given, hence the classification can be checked and validated. Experiments on WEKA show that ontology-based classification gains a relatively high accuracy and precision for classifying RSOs. When classifying RSOs with imperfect data, ontology-based classification keeps its performances, showing evident advantages over classical machine learning classifiers who either have increases of 5 % at least in FP rate or have decreases of 5 % at least in indexes such as accuracy, precision and recall.

No MeSH data available.