Topic provider


This topic is about the quality prediction of manufacturing processes. By adopting Machine Learning methods, parameters used during the manufacturing processes will be used to build models for quality prediction. The data is provided by an opto-electronic technology company. By analyzing the data, results of the manufacturing process can be accurately predicted. This can not only improve product quality, but also lower the costs of manufacturing.

The complexity of the manufacturing processes puts challenges on analyzation and optimization. Many factors will influence product quality, such as machine parameters, the amount and ratio of materials (primary materials and accessory materials), temperature, pressure, time, etc., and these factors can impact the result either directly or indirectly. Different products have different requirements, or acceptable quality level. An accurate model can help understand the relative importance of different factors and identify how the parameters can influence the quality, and thus can directly improve manufacturing process analyzation and optimization. In this topic, participants need to predict two target numbers regarding each set of parameters.



The following awards will be given to the participants who have achieved all of the following: the top three participants whose score on the Private Leaderboard is lower than the baseline (MAE < 0.007) and the report is submitted before 7/22.

First place: 150,000 discount points of hicloud + NTD 50,000 (Tax included)

Second place: 100,000 discount points of hicloud + NTD 30,000 (Tax included)

Third place: 50,000 discount points of hicloud + NTD 20,000 (Tax included)

Honorable mention: 50,000 discount points of hicloud (multiple winners, depends on the final result)


* Hicloud discount points were provided by Chunghwa Telecom

1. Hicloud points can redeem service charge. After the discount, the price will be charged at 30% off based on the list price automatically.

2. Chunghwa Telecom deserves the right to make changes to the terms and conditions herein.

3. The example of discount points calculation can be referred to https://aidea-web.tw/computing

Report content should include hardware and software specifications, data preprocessing, choice of models, and parameters adjustments, and etc.


Activity time

The activity time will be based on Taiwan Standard Time (UTC+8), and the schedule is as follows.

2020/06/12Registration begins
2020/06/19Submission begins
2020/07/15Submission deadline
2020/07/22Report submission deadline
2020/07/24Awards announcement


Evaluation Criteria

Evaluations are conducted by calculating the Mean Absolute Error(MAE)[1]

The formula is as follows:

$$MAE = {\sum_{i=1}^{n} \left| (y_i - \hat{y}_i) \right| \over n} $$

[1] Mean Absolute Error (MAE):


  • The evaluation will be based on the final uploaded result. If multiple participants get the same evaluation scores, the time of uploading will determine the ranking.
  • The maximum number of uploading times is 8 times per day.
  • This topic does not offer a team-up option, only one account per person is allowed, and each person can only participate once. If any violations are found, people who are involved would be forced to withdraw the activity.
  • When using external data sets, participants should avoid using future data as the basis of the prediction results, and must state source of data sets in the forum for references.
  • All the data, techniques and source codes are original work of participants or are used by permission complying with laws and regulations. If any third-party claims their intellectual property rights or other rights and interests are being violated, the participant will need to handle the disputes personally. If any participants violate intellectual property rights, they will be disqualified, and shall bear legal responsibilities.
  • All the achievements and their IPRs (intellectual property rights) belong to the participants, and the Copyright License Agreements, patent applications, technology transfer and equity distributions of them, should be in accordance with the relevant Laws and Regulations.
  • There should be no answer discussions between different accounts, or it will be considered as cheating.
  • If there is cheating or fraud during the activity, the participant that cheats will be disqualified from the activity and the vacancy would be filled up by other participants in the ranking order.
  • After uploading, the answers of test data would be divided into two parts to calculate the score:
    • Before the deadline of the activity: The system refers to only a part of the test data to examine and calculate the score. The result will be posted on Public Leaderboard as reference for the final score and ranking. This data accounts for 40% of the entire test data.
    • After the deadline of the activity: The system refers to the remaining test data (60%) to examine and calculate the score. The result will be posted on Private Leaderboard as reference for the final score and ranking.
  • After being notified, finalists need to submit their report before 7/22 to be eligible for the award (the report will not be disclosed or published).
  • Should there be disputes, the organizer reserves the right to make the final decision.