Added a "Comment Quality Filter" feature that further complements the existing comment block.
In addition to the conventional function that evaluates aggressiveness etc. when commenting by machine learning,
Inappropriate comments are automatically hidden by continuous analysis and evaluation, including past posts.
Streamers can go to the "Broadcast slot management page" → "Disable and limit" → "Comment quality filter" settings,
You can choose how much to filter your own distribution.
Default state (minimum restrictions):
Filtering is done to the extent that a minimum level of order can be maintained.
Hides comments from users who make multiple defamatory or discriminatory comments that could raise legal issues.
Inappropriate comments tend to be posted by the same user in large numbers, so most people are not affected at all.
Gentle world (loose restrictions):
Comments from users who post offensive comments will be hidden.
A world of angels only (high limit):
Comments from users who post comments with a violent tone will be hidden.
[Frequently asked questions and answers]
Q. Why are specific player names, organization names, country names, etc. hidden?
A. We learned about filters from past comments made on kukuluLIVE over the past few years.
Proper nouns, handle names, country names, names, etc., which tend to be inappropriately mentioned on kukuluLIVE, may also be subject to regulation depending on the situation.
They do not represent kukuluLIVE's position, nor are they set up with the intention of excluding the subjects mentioned.
It has been set up solely to learn past user trends and prevent slander, etc. against the target.
In addition, we may also set this information for protection if a complaint is received from a party regarding defamation, etc.
Q. What should I do if it becomes hidden?
A. If the filter setting by the broadcaster is set to the strictest setting, all messages that are not polite and positive will be hidden in the broadcast. Restrictions will be one-off or temporary unless the person repeatedly makes discriminatory comments.
Q. Why am I being banned even though I have not made any particularly problematic comments?
A. There are several possible causes. The strictest broadcaster filter settings will hide anything that isn't polite and positive.
In addition, in order to prevent penalties from being avoided due to IP address changes, etc., connection sources, proxies, VPNs, etc. with no history will be judged as unreliable and will be evaluated more severely.
Alternatively, they may be restricted by other features such as comment blocking, provider restrictions, etc., rather than filters.
There are cases where comments that you think are not a problem actually constitute a serious violation.
Comments are content that is publicly transmitted to the entire world using the Internet.
All comments are expected to be offensive to public order and morals at all times.
Q. Is it okay to leave it to AI? I am dissatisfied.
A. Although the operation is not perfect, we are constantly monitoring it manually and it is working as expected.
Due to the nature of penalties such as filtering, it is difficult for the target audience to understand or accept them.
It goes without saying that broadcasters, viewers, and administrators all can enjoy peacefully, but the service cannot be operated sustainably without compliance with laws and regulations and advertiser policies.
Please use the comment function only if you can help us.
Q. If I write bad words about administrators or specific people such as women, will I receive a severe penalty? Unfair?
Abusive comments to anyone are prohibited.
In addition, our service does not emphasize fairness.
On the premise of protecting the socially vulnerable, priority is given to service sustainability and technical interest, and unfair responses may be taken if it is judged to be useful for achieving them.
Automatically Translated. (Show original)