Some questions about ur word2vec detecting
Closed this issue · 1 comments
It's a good project that I interested in.
But could u explain more details about how ur team use the model to do the text extraction, I mean the files in neo4jcsv?
Waiting ur answer! Thank u very much!
The files in neo4jcsv is the import files to storage in neo4j database. Use the cypher which is the neo4j query.
And there is no text detection problem... only text classification...
For natural language query like "2017年龙湾区的冰毒数量”， After JIEBA word segmentation.
I use pre-trained word vector(also give links in README.md) model to transform the word like "年" to vector which dimension is 300.
Vertically stacked vectors combine one matrix which can used in CNN text classifier or BiLSTM, LSTM can input the vector as x through time.