Facebook has been investing heavily in Natural Language Processing to boost programmers’ productivity.
Facebook wants to enable developers to evaluate AI-based natural-language code search system efficiently. So, the company’s AI research division released a library of coding questions with matching answers.
Facebook AI also made benchmark results for many of its code-search models available to programmers. The release also included a training corpus of more than 4 million Java methods parsed from over 24,000 GitHub repositories.
This substantial investment in NLP demonstrates two things:
- Facebook wanting to expedite and streamline development
- An example of AI supporting and optimizing human labor
Let’s break it down, shall we?
Using Natural Language Processing to Boost Developer’s Productivity
Software developers sometimes run into coding problems and have to turn to Stack Overflow to learn how to solve their issues. For example, they could type “How to solve an Android bug,” and the developers’ community would answer the coding question.
While the community could help address many coding issues, it’s still far from perfect. Stack Overflow is not always useful for developers working on a less-common programming language or using proprietary APIs.
That’s where Facebook’s Neural Code Search comes in.
It’s a machine learning method for training neural networks to answer “how-to” coding questions. In this case, the source code itself would serve as training data, which generates NLP systems to answer coding questions.
Programmers can type input their queries, and the system would learn from the source code to provide an answer.
A Reddit user wrote:
“[O]ften times the most difficult part of programming is to describe exactly what I want to do in concise natural language.”
Facebook’s goal is to address this issue. The tech company intends to use the NLP research to improve software developer’s productivity.
Besides, the effort also indicates how we can use artificial intelligence to support and optimize human labor. Instead of stealing our jobs, AI could make us better at it.
MY PERSONAL EXPERIENCE ON LOST CRYPTO RECOVERY!
I read so many stories about bitcoin loss to scams. I will like to start by saying the agencies responsible for bitcoin security has really done nothing to help locate stolen or lost coins. In my situation my MacBook was hacked by someone that had access to my emails, i immediately contacted blockchain and they only wasted my time, after which i worked towards getting help else where, i was referred to consult a bitcoin expert who helped track and retrieved my 3.3 btc, for an agreed fee. I was more than grateful and willing to pay more after the job was done. Thankful i didnt fall victim and would like to recommend ( MORRIS GRAY 830 @ G maiL . COM )