YouTube算法如何工作？

With over a billion users and billions of hours of video, the fact that YouTube’s algorithm manages to deliver what you want to watch when you visit the site is a testament to software engineering. So, how does it work?

YouTube的算法能够吸引超过十亿用户和数十亿小时的视频，从而能够在您访问网站时提供您想要观看的内容，这一事实证明了软件工程。 那么它是怎样工作的？

The short answer: Nobody knows the details—not even YouTube, to an extent. YouTube’s algorithm uses machine learning to suggest videos, which means there are no set rules we can tell you. Besides, Google wouldn’t tell us anyway, as that would lead to people exploiting them.

简短的答案：在某种程度上，没人知道细节，甚至连YouTube都不知道。 YouTube的算法使用机器学习来推荐视频，这意味着我们没有可以告诉您的固定规则。 此外，Google不会告诉我们，因为这会导致人们利用它们。

我们所知道的 (What We Do Know)

When you train a machine learning model, you give it a bunch of input and then rank its suggested outputs on how right they are.

训练机器学习模型时，您需要给它一堆输入，然后根据其正确性对建议的输出进行排名。

Here’s a greatly oversimplified example. Say you wanted to train an AI to tell the difference between pictures of cats and dogs. Essentially, you’d give an AI a bunch of pictures of cats and dogs, have it start choosing, and then score it right if it answered correctly. The more it gets correct, the better it gets at choosing. The result is a machine that can identify cats and dogs. This training uses a metric by which results are judged; in our case, the cat-o-meter, or what percent of the image is indeed cat.

这是一个大大简化的示例。 假设您想训练AI来分辨猫和狗的图片之间的差异。 本质上，您会给AI提供一堆猫和狗的图片，让它开始选择，然后如果回答正确就给它打分。 它越正确，就越适合选择。 结果是可以识别猫和狗的机器。 该培训使用一种指标来判断结果； 在我们的例子中，猫眼图或图像的百分比确实是猫。

The metric YouTube uses is watch time—how long users stay on the video. This makes sense because YouTube doesn’t want people skipping around looking for videos to watch, as that’s more work on their end, and less time spent watching.

YouTube使用的指标是观看时间，即用户观看视频的时间。 这是有道理的，因为YouTube不想让人们无所事事地寻找要观看的视频，因为这样做可以使他们的工作更多，而花费的时间更少。

It’s much more nuanced than just “how long you watched a video,” though. The algorithm takes into account many different factors and ranks them accordingly: viewer retention, impressions to clicks, viewer engagement, and some other behind the scenes factors that we never see. YouTube then tailors these factors to your profile so that it can suggest videos you’re more likely to click.

不过，这比“您看了多长时间的视频”要细腻得多。 该算法考虑了许多不同因素，并对它们进行了相应的排名：观看者保留率，点击印象，观看者参与度以及我们从未见过的其他一些幕后因素。 然后，YouTube会根据您的个人资料调整这些因素，以便为您推荐更有可能点击的视频。

从中拿走什么 (What to Take Away From This)

If you’re an aspiring YouTuber, the two main things to work on are maximizing your average view duration, and maximizing your click-through rate. Take the following upside-down pyramid.

如果您是一位有抱负的YouTuber，那么需要做的两件事是最大程度地延长平均观看时间，并最大程度地提高点击率。 采取以下颠倒的金字塔。

YouTube suggests your video to a bunch of people, on the home screen and in the suggested tab. On my account, I have almost 750 thousand impressions. That seems pretty good, but only a fraction of those people click your video. This fraction is called your click-through rate, and it’s measured as a percent (you can see in my example that I have a 4.0% click-through rate). The Views figure shows the actual number of people that clicked through.

YouTube在主屏幕和“建议”标签中向一群人推荐您的视频。 在我的帐户上，我有将近75万次展示。 这看起来不错，但只有少数人点击了您的视频。 这个分数称为您的点击率，以百分比来衡量(您可以在我的示例中看到我的点击率为4.0％)。 “观看次数”图显示了点击的实际人数。

After someone does click the video, YouTube then measures the amount of time those people spent watching the videos.

有人点击视频后，YouTube会衡量他们在观看视频上花费的时间。

You can see why so many YouTube creators use clickbait titles and thumbnails (to get those click-throughs) and long, drawn out videos (to up retention time). These are two very annoying traits of many YouTube creators, but hey, blame the algorithm.

您会看到为什么这么多YouTube创作者使用clickbait标题和缩略图(以获得这些点击)以及冗长而引人入胜的视频(以延长保留时间)。 这是许多YouTube创作者的两个非常烦人的特征，但是，请怪这算法。

案例研究 (A Case Study)

()

Let’s take a look at two big channels that take different approaches to tackle the algorithm. The first is Primitive Technology, a channel run by a guy who goes into the wilderness and builds things with no tools. All of his videos are very long but keep up a good level of engagement throughout that length—quite an accomplishment as there is no narration. This fact means that he probably has a very high average view duration, which is good in the algorithm’s eyes.

让我们看一下采用不同方法解决算法的两个主要渠道。 首先是原始技术，这是一个由一个人经营的渠道，这个人走进旷野，不用工具就能建造东西。 他的所有视频都很长，但是在这段时间内保持了很高的参与度，因为没有旁白，这很不错。 这个事实意味着他可能具有很高的平均观看时间，这在算法的眼中是很好的。

Because he only makes one video a month, it’s surprising that he has over 8 million subscribers. This is probably because the long time between videos creates a feeling of something new when the next one drops. His videos are iconic, and whenever they show up in my feed, I almost always click them. I’m guessing others feel the same way, so he probably also has a high click-through rate as well.

由于他每个月只制作一部视频，令人惊讶的是他拥有超过800万的订阅者。 这可能是因为视频之间的长时间间隔会在下一个视频掉落时产生新的感觉。 他的视频具有标志性，每当它们出现在我的Feed中时，我几乎总是会点击它们。 我猜想其他人也有同样的感觉，所以他可能也有很高的点击率。

The second channel takes a slightly scummier approach. BCC Trolling, a Fortnite “Funny Moments” channel, takes clips from popular streamers and edits them into daily videos. In the last year they’ve mastered the algorithm and shot up to 7.3 million subscribers. To maximize watch time, they put the title clip of the video somewhere in the middle of the video, forcing people to watch it for a while before coming to the clip they clicked on, essentially getting them “hooked” on the video. Because of this, their watch time is higher.

第二个渠道采取了一些卑鄙的做法。 BCC Trolling是Fortnite的“有趣时刻”频道，它从受欢迎的彩带中提取片段并将其剪辑为每日视频。 去年，他们已经掌握了该算法，并吸引了730万订户。 为了最大限度地延长观看时间，他们将视频的标题片段放在视频中间的某个地方，迫使人们观看视频一段时间，然后再单击所单击的片段，从而使他们“钩住”了视频。 因此，他们的观看时间更长。

They’re also excellent at clickbait thumbnails and titles, putting *NEW* in all caps on many videos, and always with colorful thumbnails that are usually custom-made, and often very misleading. But, they’re not obvious clickbait; the videos do deliver on the title, but it’s just clickbait enough to get people to click.

它们在点击诱饵缩略图和标题方面也很出色，在许多视频的大写字母处都加上了* NEW *，并且始终带有通常是定制的彩色缩略图，并且通常会产生误导。 但是，它们不是明显的点击诱饵； 视频确实会显示标题，但点击诱饵足以吸引人们点击。

This is the main thing to take away from BCC: if you’re going to clickbait your thumbnails, do it subtly. Putting outright lies in the title will often make people angry and may have the opposite effect you intend.

这是摆脱BCC的主要目的：如果要单击诱饵缩略图，请巧妙地进行操作。 在标题中摆出绝对的谎言通常会使人生气，并可能产生与您预期相反的效果。

Either way, you should find what works for you, and use that to your advantage. Keep watch time and click-through rates in mind going forward, but stick to your format, and don’t let the algorithm dictate your content.

无论哪种方式，您都应该找到适合自己的方法，并利用它来发挥自己的优势。 请牢记观看时间和点击率，但是请遵循您的格式，并且不要让算法决定您的内容。

翻译自: https://www.howtogeek.com/364720/how-does-the-youtube-algorithm-work/