In order to show the effectiveness of using (a) library loan records and (b) information about book contents as a basis for book
recommendations, we entered various data into a support vector machine (SVM), used it to recommend books to subjects, and
asked them for evaluations of the recommendations that were given. The data that we used were (1) confidence and support
with an association rule that was based on the loan records, (2) similarities between book titles, (3) matches/mismatches
between the Nippon Decimal Classification (NDC) categories of the books, and (4) similarities between the outlines of the
books in the BOOK Database. The subjects were 32 students who belonged to T University. The books that we recommended
and the loan records that we used were obtained from the T University Library. The results showed that the combinations of (1),
(2), (3) and (1), (2) were rated more favorably by the subjects than the other combinations. However, the books that were
recommended by Amazon were rated even more favorably by the subjects. This is a topic for further research