By Xiaodong Gu, et al.
Presented by David A.N
Approach | Technique | Novelty |
---|---|---|
Sourcerer (Linstead et al., 2009) | IR | Combines the textual content of a program with structural information. |
Portfolio (McMillan et al., 2011) | PageRank | Returns chain of functions through keyword matching. |
CodeHow (Lv et al., 2015) | IR | Combines text similarity matching and API matching. |
Approach | Technique | Novelty |
---|---|---|
Sourcerer (Linstead et al., 2009) | IR | Combines the textual content of a program with structural information. |
Portfolio (McMillan et al., 2011) | PageRank | Returns chain of functions through keyword matching. |
CodeHow (Lv et al., 2015) | IR | Combines text similarity matching and API matching. |
Not Machine Learning
Approach | Novelty |
---|---|
White et al., 2016 | Predicting software tokens by using RNN language model. |
DEEPAPI (Gu, et al., 2016) | Deep learning method that learns the semantic of queries and the corresponding API sequences. |
Source code and Natural Language queries are heterogeneous
They may not share common lexical tokens, synonyms, or language structure
They can be only semantically related
Query: "read an object from an XML"
Can anybody see the problem?
A sentence can also be embedded as a vector.
2 ways for embedding:
CoNN
CoNN
DeNN
The hinge loss
Parameters of NN
Training Tuple
Constant Margin
Embedded Vectors
Preprocessing as camel-case and AST parsing.
Source: GitHub and JavaDoc
Preprocessing as camel-case and AST parsing.
Source: GitHub and JavaDoc