Language of courses:
English
Introduction in chinese:
This. Style. Width=600;”
In some areas of application, such as sign language, medicine and sensor networks, incidents may not necessarily be instantaneous, but they may take some time. Spacing-based sequences of events may contain useful field knowledge; therefore, searching, indexing and digging for such sequences are essential. This paper presents two distance measures for comparing inter-temporal sequences of events that can be used for multiple data mining missions, such as classification and clustering. The first measure maps each event sequence based on spacing to a set of vectors that contain all simultaneous event information. They are then compared using existing dynamic programming methods. The second method, known as the artemis, is to find a correspondence between the intervals by mapping the two sequences into two maps. Similarity is inferred from the hungarian algorithm. In addition, we have proposed a linear lower time limit for artemis. The performance of both methods is tested on data in three areas: sign language, medicine and sensor networks. Experiments have shown that halogens are superior in terms of high-level artificial introduction of noise。
Course introduction:
In all applications, such as sign language, media, and internet networks, events are not necessarily efficient, but they can have a time historyI'm sorry, but i'm sorryWe understand two differences for competing considerations of interval-bI'm not sure if i'm going to tell you, which could be used for some sort of catalog, such as classification and reportingAs long as i'm here, i'd like to ask you a question, but i'd like to ask you a questionI don't know, out all coThe second meeting was called artemis, and it was a big dealIn addition, we present a linear-time low for artemis.
Key words:
