发明名称 |
SYSTEM, METHOD, AND RECORDING MEDIUM FOR VECTOR REPRESENTATION OF WORDS IN A LANGUAGE |
摘要 |
A method, system, and non-transitory compute readable medium for vector representation of a sequence of items, including receiving a sequence of items from a source, producing a first distributed representation for each item of the sequence, wherein the distributed representation comprises a word vector and a class vector, partitioning the sequence of items into classes, and training the received sequence using the first distributed representation, such that a new distributed representation is produced for which the vector entries of the new distributed representation are amplified when the vector entries of each item correspond to a class of an item to be explained and fractionalizing vector entries of each item that do not correspond to the class of the item to be explained. |
申请公布号 |
US2017109648(A1) |
申请公布日期 |
2017.04.20 |
申请号 |
US201514886167 |
申请日期 |
2015.10.19 |
申请人 |
International Business Machines Corporation |
发明人 |
Shmueli Oded |
分类号 |
G06N99/00;G06F17/28 |
主分类号 |
G06N99/00 |
代理机构 |
|
代理人 |
|
主权项 |
1. A method for vector representation of a sequence of items, the method comprising:
receiving a sequence of items from a source; producing a first distributed representation for each item of the sequence, wherein the first distributed representation comprises a word vector and a class vector; partitioning the sequence of items into classes; and training the received sequence using the first distributed representation, such that a new distributed representation is produced for which the vector entries of the new distributed representation are amplified when the vector entries of each item correspond to a class of an item to be explained and fractionalizing vector entries of each item that do not correspond to the class of the item to be explained. |
地址 |
Armonk NY US |