ABSTRACT
A large amount of high-precision information about celestial bodies such as the position,
motion and parallax of stars are recorded in the Ibarco satellite. As the nearest open star
cluster to the earth, finding out the member stars of the Bi star cluster and drawing the Hero
diagram is important in cosmological research. This paper mainly establishes the OPTICS
clustering model based on the improved DBSCAN algorithm, finds out the member stars of
the Bi star cluster according to the distance, direction and location of the stars, and draws the
Herro diagram of the Bi star cluster.
In response to question 1, it is required to confirm the member stars of the Bi star cluster
in the data set. First, we performed a descriptive statistical analysis of the attached data, and
found that the standard deviation of the data of the star’s right ascension and declination and
the self-property component was large, so we carried out the normal distribution test on the
data of these four variables, and used the Laida rule to eliminate the degree of dispersion.
Larger data. Then we standardize the variable data and choose the density-based DBSCAN
clustering method. In order to reduce the sensitivity of the DBSCAN algorithm to the radius
and threshold of the initial parameter domain, we improve the DBSCAN algorithm and
introduce two concepts: core distance and reachable distance , Using the OPTICS clustering
model and clustering according to the distance, direction and location of the stars themselves,
331 member stars of the Bi star cluster are obtained. Finally, the member stars of the Bi star
cluster are verified on the basis that the Bi star cluster satisfies the globular shape.
In response to the second question, it is required to draw the Herro diagram of the
member stars of the Bi star cluster. First of all, we learn about the relevant information of
Herrotium by consulting the literature. Then we calculate the absolute magnitude of the star
based on the two data of the apparent magnitude and the distance of the star from the
observation point, and obtain the vertical axis data of the Herrot chart, and use the color index
given in the attachment as the horizontal axis data. Finally, we draw the Helo diagram of the
Bi star cluster by using the standard Helo diagram in the comparison literature of Matlab and
PowerPoint.
Finally, the model proposed in the paper is deeply studied and analyzed, the advantages
and disadvantages of the model are pointed out, and the ideas for improvement of the
optimization of the model are proposed.
Keywords: Hyades; Raida Criterion; Improve DBSCAN algorithm; OPSTIC clustering
model; Herotu
