Preface
Hello everyone, I am Evan Lin from the LINE Taiwan DevRel team. I am very happy to share with you the third developer meetup of this year. This is also the first offline meetup held in Hsinchu after the pandemic. It is also the first time the LINE Taiwan engineering team has come to National Chiao Tung University to hold an offline event.
KKTIX event webpage: Event URL
At this LINE Developer Meetup, first, LINE Taiwan CTO Marco Chen introduced the LINE Tech Fresh Internship Program, and then Shawn Tsai, the head of the LINE Data Engineering team, brought "How ML Powers LINE Services".
Article List
- Article 1: [Conference Notes] 2020/09/18 LINE Developer Meetup 13 (1)
- Article 1: [Conference Notes] 2020/09/18 LINE Developer Meetup 13 (2)
- Article 1: [Conference Notes] 2020/09/18 LINE Developer Meetup 13 (3)
LINE TECH FRESH (LINE Technical Star Talent Program) Student Internship Program Promotion / CTO Marco Chen
The first to take the stage was Marco Chen, the CTO of LINE Taiwan. As the head of the LINE engineering team, Marco explained to all the students the reasons for the establishment of TECH-FRESH. Here are some excerpts from the content of that day:
To the guests who are currently employed, please raise your hand if you were thinking about doing your current job when you were in school.
To the students in school, please raise your hand if you are very sure what your job will be in the future.
Do you know how many professional roles are involved in developing a high-traffic system at LINE or a company like LINE that provides internet application services?
How many technologies are used? Do you know what roles and positions are there?
TECHFRESH is a technical internship program for students at LINE Taiwan with two main purposes:
- To help students who are preparing to enter the workplace to truly understand what software development jobs are available in an internet application service company. This can allow students to find the job content that suits them best in their careers and develop their expertise. Regardless of whether they develop at LINE in the end, we also help this country cultivate good talents who can contribute the most to this country.
- To cultivate talents familiar with our development technology and processes in the process, and then be willing to stay at LINE to develop. The original LINE summer work-study program, which lasted for three months, was too limited for students in information science to really understand all the roles and job content in the system development life cycle. Therefore, we proposed a one-year internship program to the headquarters, coming three days a week, and the work content is led by TPM, performing some one-time project development, or joining the project team to support the development work of the project team.
Through this conversation, students can clearly understand the main reasons and origins of the establishment of LINE TECHFRESH. Through a one-year internship, students have the opportunity to truly learn cross-border product development cooperation and software engineering development experience. I hope students will sign up quickly! Application URL.
Related Information:
- LINE TECH FRESH – Technical Star Talent Program
- LINE TECH FRESH – Technical Star Talent Program, Internship Experience Revealed
How ML Powers LINE Services / LINE Data Team - Shawn Tsai
The second to take the stage was Shawn Tsai from the LINE Taiwan Data Engineering team, who shared with everyone how machine learning can make LINE's services more user-friendly.
Composition of the LINE Taiwan Data Engineering Team
First, Shawn shared with everyone the composition of the LINE Taiwan Data Engineering team, which is mainly composed of the following three roles:
Data Engineer
As a data engineer, you need to have strong engineering skills, whether it's data extraction, grabbing, and pre-processing. Even the data exploration part, and finally the deployment of machine learning models, cannot be done without the assistance of data engineers.
Data Scientist
The job of a data scientist is to assist data engineers in extracting data and discussing how to pre-process it. Then, the machine learning model is learned.
Data Analyst
The focus of data analysts is on data exploration, finding the values that can truly solve the problem. And make relevant tests and corrections to the completed model.
Cooperation between the Data Engineering Team and Projects
The Data Engineering team is mainly composed of the above three roles. All data engineering teams will have different task groups due to different product needs. Some products are still in the data discussion and extraction stage, and some products may have entered the machine learning model tuning stage. For different product lines, each member can participate in many interesting products and projects in their daily work, and can learn new machine learning model methods to apply to each daily work.
Challenges Faced by the Data Engineering Team
Because LINE has more than 21 million users, LINE TODAY produces one million articles a year, and LINE Shopping has five million product inquiries every month. So much data is the problem that the data engineering team has to face. And machine learning itself can be simple or quite complex. Next, the machine learning techniques used will be explained slightly according to the different products.
LINE Customer Service Assistant
"Oh no... how do I move my account when I change phones?"
"How do I buy stickers to send to friends and family?"
These questions are operation questions that users want to know every day, but how can they find answers in a timely manner? At this time, you can use the "Customer Service Assistant" machine learning capabilities to help you reply quickly. Mobile phone Click here to join LINE Customer Service Assistant account or search for @linehelptw to add friends. For more usage introductions, please refer to "LINE Customer Service Assistant" smart customer service has been upgraded~ Solve LINE problems in conversation.
Because the same question may have various ways of asking, for example:
- Why sometimes the notifications don't come?
- Why can't the messages come out?
- LINE doesn't ring or vibrate?
These three completely different ways of asking may lead to the problem caused by the iOS 11 update at the time. It takes a lot of time to reply to these manually, so you need to use Natural Language Understanding (NLU), and use LSTM to understand the relevance of the text in the context, and use CNN to obtain the features of the text and other texts. This is the first version of the solution, but the effect is not satisfactory. Later, through the use of seq2seq, CBoW, DSSM and BERT to achieve the Esemble solution, this method has greatly achieved better results.
LINE Message Verification Assistant
The "LINE Message Verification" platform was officially launched last July. It not only has an official website, but also connects to the LINE official account. Users only need to "forward" the messages received in the chat room to the "LINE Message Verification" official account. If there are already verification reports in the database, the verification assistant will automatically determine its authenticity, and the system will provide the verification results immediately; if the message has not been verified, it will be reported to a professional verification unit, and after clarification, it will return the correct information to the user as soon as possible, providing the most immediate message identification service, and assisting users in identifying the authenticity of suspicious messages, reducing the chance of false messages being spread again.
The number of messages received every day is as high as 40,000, but manual identification can only handle 300 messages per day, so machine learning is needed to help a lot at this time. Through Near-Duplication and Classification, two methods are used to find and classify messages. Now the efficiency of message verification has improved more than ten times, and 46% of the suspicious messages sent by users have been successfully clarified. Don't want to become a spreader of false messages? Join the "LINE Message Verification" official account.
Learning through Machine Learning Projects
Many student friends are curious about what exactly the data engineering team does every day when working at LINE? The speaker also generously shared with everyone that most of the time every day as a data engineering team is really spent on training machine learning models? This picture can let you know that most of the time is spent on machine settings (Configuration), data collection (Data Collection), and a lot of time on data verification (Data Verification), and even the construction of the infrastructure (Serving Infratructure) that can continuously update the latest machine models after the model is built will also take a lot of time. The actual model training is often only a small part of the entire project. You can also know that the main time and expertise of data scientists lie in how to find and identify "key information".
Related Articles:
About the "LINE Developer Community Program"
LINE launched the "LINE Developer Community Program" in Taiwan at the beginning of this year, and will invest long-term manpower and resources in Taiwan to hold developer community gatherings, recruitment days, developer conferences, etc., both internally and externally, online and offline, and has already held more than 30 events. Readers are welcome to continue to check back for the latest updates. For details, please see:
Top comments (0)