In this article, I will visualize and analyze Youtube Turkey Trend data that I have collected for 10 days. We will visualize data and make inferences about what it tells us.

The highest rungs of Turkey YouTube Trends generally seem to follow the World youtube trends. Because the video in the top 10 was in Trends in all youtube in general. By looking at this, we can understand that Turkish Youtube users are not far from global trends. Despite the barrier of language, culture and life, we can say that Turkish users are interested in content that is popular in the rest of the world. But there is a reason for this trend. Shorts. Youtube Shorts videos are consumed with a UX similar to Tiktok. Considering that the Turkish user is a fierce scroll-watching fan, it is inevitable that the most watched videos of Youtube Turkey trends are always foreign shorts videos, Turkish shorts videos are also present in the trends. Both the shorts’ support for global content consumption and its easy consumption (the algorithm suggests automatically, the videos are short and there is no search-finding problem). Other videos other than shorts videos are generally popular on social media (the rap song that Shakira buried her cheating husband in the ground. The song was heard on twitter and instagram and people may have come to youtube to listen.) or youtube is one of the mega content producers of the global. videos (like MrBeast. The videos of these creators are always trending on Youtube Global because they are the producers that Youtube has the most subscribers and advertising revenue.). In addition to this, K-Pop and Korean-based content also found its place in Turkey Trends. After the age of 12 in Turkey, South Korean love is automatically imposed on our young girls, so this seems normal to me. You can see the popularity of South Korea among Turkish youth by scrolling a little on Social Media (especially on TikTok). In addition, there are shorts videos in Other Languages (Russian and Vietnamese), the whole reason for watching them is that the content is interesting, otherwise it is difficult for a normal level creator to overcome this language barrier otherwise. Turkish TV series seem to have found themselves in trends. Especially the TV series that have gained popularity on Twitter and Instagram lately, such as the Kingfisher, have thrown himself into the trends of youtube. But of course, if this number of views does not come from the number of global views, that is, all of them are Turkish users, we will also look at the contents that are rooted in Turkey or Turkish. Let’s see who or what is trending in Turkish content.

When we take a look at the Turkish content in Youtube Turkey trends, there is almost complete dominance of Series, TV Shows and Music. Almost every episode of the Kismetse Olur program seems to be trending as soon as it comes out. In addition, the newly released series have found their place in the trends, and the popular TV series of recent times are at the top. Trend music, on the other hand, has taken its place in the trends with the popularity coming from tiktok. Internet programs are also included in the trends. For example, Topics:Open Mic. Each episode of the program, in which Oğuzhan Uğur summoned political leaders and asked challenging questions with intriguing answers, ensured that he could enter the trends. In other words, to summarize in general terms, besides the dominance of TV Shows, Series, Internet Programs, Sports Channels and Music, Vlog channels and Challenge content channels also have a presence. If you ask me, the popularity of instagram and tiktok has a lot to do with Turkish content trends. There are users who come to listen to popular songs on Tiktok, there are those who come to review the content they see in their instagram edit-posts in detail. I can say that the reason why even these series are so popular is the edit videos published with the sparkly and mouth-watering titles on Instagram. So if you want to be in YouTube trends, I think you need to become popular on Social Platforms. Otherwise, there are other ways to get a lot of views or mass, but what I mean is that YouTube trends usually follow the popularity of social media, but of course, there are situations that deviate from my thesis. My opinion is simply that what is popular in one place is easy to become popular in another.

When we take a look at the countries where the channels in Youtube Turkey Trends produce content, Turkey is normally ahead. In addition to this, the contents of US-based channels are the most found in trends after Turkey. I see this because US-based channels produce more global content. Of course, another important point in this is what kind of videos are watched, we will see them. Once again, we observe the place of Korean-based content in Turkey’s trends. In addition, content from channels broadcasting from other countries, albeit sporadically, was able to find a place in Turkish trends.

In Youtube Turkey trends collected in 10 days, the rate of Youtube Shorts is seen as 15.42%. This is a big number. But besides that, I also want to look at the shorts ratio of the 100 most watched trending videos.

38 percent of the videos that are in the top 100 in Youtube Trends are seen as Shorts videos. Shorts from the Youtube Content Production ecosystem seems to have taken its place immediately, and it is not difficult to understand why. It is a content that is not difficult to produce, distribute and consume. Since it is a type of content that is generally watched without much thought and is fast, does not cost much to the content producer, and is easy to distribute (such as sharing, editing), as well as the emergence of TikTok, the trend to consume content by swiping and short has become popular. normal. But one thing I noticed in youtube shorts is that they are mainly short versions of videos published on youtube. Producers serve as shorts in their long videos by turning points of interest and hype points into a short video. It can be said that this basically contributes to the consumption of feature length videos. The interest of the user in the shorts is tickled and he is allowed to watch the feature length video. For most channels on YouTube, Shorts is seen as a kind of trailer. I think YouTube is turning into a television. Feature-length content, Trailer and Advertising within Content. The pattern is almost the same.

When we examine the categorical distribution of the videos in Youtube Turkey Trends, it seems that Entertainment videos dominate. The Sports category, which generally includes match summaries and match comments, comes right after, as well as the Film/TV Series/Animation and Blog categories, but I was surprised that the Blog category was higher than the Music category, in fact, I think it was due to the speed of content production. The difficulty and frequency of the production of music content and daily life content are not the same, besides, the areas where the music category is shared are higher than the blog category. People generally listen to music on Spotify or other platforms in Turkey. But of course, this still does not change the place of music in Youtube Turkey trends. I wondered why the blog category was so high. Let’s also do content analysis on the basis of categories.

In the Charts above, the average number of views, likes and comments of the Trending Videos are listed on a categorical basis. While the Comedy Category is high in Views and Likes, the Music category seems to have made a difference on the basis of Comments. The Entertainment category, which produces the most videos on the train, seems to keep average figures in the statistics. We can say that the blog category has kept a statistic parallel to the number of videos in the statistics. I think that the fact that the blog category has such a high number of likes despite this number of views is the reason why the audience is more interested in the person who publishes the content rather than the content. The audience shows likes no matter what the video of the influencer they love. The Sports Category, which is one of the categories that produces a lot of videos on the train, is in an average position in the statistics. Already, when I took a look at Youtube sport videos, I saw this. While short videos such as Highlights have a high viewership rate, videos such as comments and analysis receive a low viewership compared to summaries. The reason for this is understandable. The audience who cannot watch the match shows high interest in the summaries. I think that the reason why the comments are so incomprehensible is the size of an audience that follows sports in such detail. The general public follow their teams, watch the match highlights and see them live in the stadium a few times a year, pick up their jerseys and scarves. But the majority don’t care much about the performance of the players, the analysis of the last 5 matches or the deep performance of the opponent. With such matters, they make fans, which we can call more fanatical. Therefore, while the statistics of the summaries are high, the comment-analysis content, which is a more detailed content, keeps more average statistics. However, the sports category is strong, especially football. It is an extremely strong social phenomenon in Turkey. In addition, there is a category that ranks high in the statistics, although it cannot show itself in the number of trending videos. How-to videos. How-to videos have extremely high views, while the number of comments and likes is average. When people want to do something, like repairing a tool, installing something, they just watch. Giving likes is missing at that moment, and it may not be necessary to comment. People of this type usually seem to be watching and passing by. Category audiences really have different tendencies. Another strange category is the Gaming category. I wondered why the views and likes are lower than other categories and the gaming category is at the top of the comment numbers. People of this type usually seem to be watching and passing by. Let’s look at the ratio of views, likes and comments to each other to examine the trends of the categories on statistics.

When we look at the average Likes-Viewing rates of the categories, we see a very different picture than other tables. What underlies the Likes-Comments table is how much of the video views content producers can convert into likes. In other words, it is how many people like the video by making them appear in the future. Because if the user likes a video, other videos from that producer will be suggested by the algorithm. Dislike is a powerful parameter in the algorithm. In the chart above, we look at the average likes-to-view ratios of trending categories. In all the charts above, although the rankings changed, the categories were the same, but here we see that the categories that we never noticed were successful in converting views into likes. Categories (such as Gaming, Education, Pets, Science), which are generally at the bottom of the Average Views and Likes numbers, are quite successful in converting the views they get into likes compared to other categories. Categories with high views, such as Entertainment, Comedy, Music and Sports, lagged behind in converting these views into likes. It is necessary to understand that Converting views to likes and getting high likes are separate. You can get a high number of likes from other videos (like the Comedy category), but if this number of likes is not even 1 percent of your views, then you won’t have much success in getting likes. Because if you can’t make millions of people who watch you like the video, you can’t give them that liking. It means they’re watching you.

When we look at the average Comments-Views rates of the categories, Gaming is at the top again, the correlation of statistics between each other seems very strong in the Gaming category. In other words, the tendency of the viewer to like and comment on the video is higher than the other categories, but it is more numerically different because, as can be seen in the graph above, the average rate of even the highest category is below 1 percent. The News-Policy category, which is always at the bottom of the other charts, showed an average success in converting the views into comments. In the Education category, content producers seem to have achieved a slightly better number than other categories in converting their views into likes, as well as a good figure from other categories here. And besides, the Pet category, which was very successful in converting its Views to Likes, took the last place here, I think the main reason for this is that the videos in this category are short and the content does not lead to anything but laughter. People see a short cat video in both the shorts and the main content area, they laugh and they go through the hearts (likes), but they don’t comment much because the tendency to comment requires something different. It is necessary to investigate the necessities of the actions. Does laughing at something get people to comment on that content, or does it matter when and what you laugh at? Why doesn’t the commenter on gaming videos do this on pet videos? For this, we need to go deeper, and we will do that later.

When we look at the average values, we see the weight of the News-Policy category. Except for this category, the categories with the highest average comment-like rates are almost the same as the categorical ranking of the Views, as we said above. In the News & Politics category, I wondered why it’s in such a position here, even though it’s so low or average on other charts. Let’s take a look.

As we already know, this category has low views, likes and comments, but it is really different that it has such a balanced number of likes and comments. I think it’s because of the mass. In the future of the notebook, we will also make an analysis on the basis of the comments of the videos, and this situation will be revealed a little more. I think that the low number of likes is due to the low tendency of the audience to like, the average age of this audience is generally high. That’s why they don’t like the videos (that doesn’t mean they don’t like it, they just don’t react, they don’t press the button), they just watch it, if a certain emotion and thought arouses, they write comments, I can already guess that half of these comments are praise, half are questions. Likewise, most of the videos from this category in trends are toki draws, while others are political events and cat and dog fights of politicians.

In the data above, we revealed the correlations between parameters. The parameters with the strongest statistical strength with each other emerged. Viewing and liking data have the strongest bond with each other. Channel Tags and Categories of Trending Videos are also compatible with each other. And one thing that caught my attention was the correlation between the number of comments and caption data. Even if the subtitle is a little bit, it may cause comments. We’ll look into that. The number of Likes and Views also has a high correlation with the number of Subscribers of the Channel. In other words, we can make a semi-correct thesis by saying that the videos that have less views have few subscribers and the videos that have more views and likes have many subscribers. It should not be forgotten that trends are not determined by the size of the figures, there are other parameters. There are those who have bad statistics and trend, high and not, this is completely dependent on the YouTube rating algorithm and is something they keep secret. When we look further down the table, we see that there is a slight correlation between the subtitles and the number of subscribers of the channel. In order to be a large and globally trending channel, localization is a bit important. At least a subtitle can change a lot of things. This means that regular content is important, as the number of views of the channel correlates well with the total number of videos. Comments seem to have a bigger impact on channel subscriber counts than likes and views. In addition, there is a link between Channel Tags and the Country and Languages of the Channels, which means that certain topics are more popular in certain countries and languages. In particular, Turkish content producers generally produce a lot of challenge content, can we say that this place puts Turkish content producers in a mold in the Entertainment category?

When we examine the average video durations on a categorical basis, we see that the news and policy categories have the longest videos. This is a normal situation. The content on News and Politics is really long, especially channels such as CNN and A-Haber produce almost 6 hours of programs. In addition, when we take a look at the categories that are leading in view, like and comment statistics, the Entertainment category took an average of almost 1 hour, and the fact that the Science and Technology category took such a short time shows us that the main content type of this category is Shorts. (Youtube 60 seconds. and automatically display shorter content to users as shorts). The game category took an average of 45 minutes, I think the main reason for this is that Turkish Youtube Gaming Content producers are mostly formed around publishers. Broadcasters play games in their streams and present it piecemeal on YouTube. Since these publishers are also popular, it is possible for them to get into trends. Videos such as PUBG and LoL may be the reason why this content type has such a high average time due to the long clash times of the games, because these two games are almost everywhere in the Turkish youtube content category and these two games have weight in the trends. The average duration of PUBG Battle, one of these games, takes almost 30 minutes, and the situation in LoL is also a bit longer in games like these, where the same players fight. It can be shown among these reasons, but in my opinion the most important reason is that the manufacturer has extended the period in order to advertise more. The average video duration of the Comedy category with the highest views seems to be 10 minutes and above, the reason why they keep it above 10 minutes is that youtube does not allow advertising for videos of 10 minutes or less, I guess, these videos will be shorter. Music videos, on the other hand, take 4.5 minutes on average, while the Blog category, which still obtains good statistics, takes an average of 10 minutes, and the How-To and Style category takes an average of 12 minutes. I didn’t understand why the Series and Movie category took so long. There is an average of 38 minutes of video length. For some reason, as if they were broadcasting the movie itself, let’s take a look.

Yes, when we look a little bit, the trends usually include full episodes of trending series and upcoming episode trailers, as well as short shorts videos on various topics from foreign content producers. We can say that the main factor that raises the average so much is the full episodes of Turkish TV series, because one of them takes almost 2.5 hours.

When we look at the duration of the top 10 videos, almost most of the videos are under 1 minute and almost none of them exceed 3 minutes, except for one video. Let’s also examine the duration of Turkish content that is watched high in trends. We can say that foreign content in Turkish trends is generally shorts or videos shorter than 3-4 minutes, where the content is understood by watching. But of course, videos of content producers like MrBeast, who have come to a place in the global world, find their place in Turkey trends, no matter how long they last due to both voiceover and subtitle support. Let’s not forget that MrBeast’s channel has Turkish voiceovers.

When we examine the relationship between the duration of the video and other statistics, it can be seen that there is not much of a relationship between them. So, in trending videos, the duration of the videos does not have much effect on the statistics of views, likes and comments, besides, I would like to examine the video duration with other data. But before that, we’re going to do a few more things.

In the chart above, the total number of videos of the channels in the trends is given. When we take a look, we see the weight of news channels and TV channels, and after their leadership, popular series come, especially the fact that they can produce so many videos even though some series have just started, which shows how important regular video production and broadcasting is for statistics and development. Giving the audience something to be constantly interested in is especially important for TV channels like this, or the environment where the TV shows are constantly coming out may lose interest. Football Clubs and Football related channels also rank high in content production. To explain in general, the production of content in channels based on News Channels, TV Series / Show programs and Football is very intense, but of course, the release dates of these channels are also important, so it should be evaluated as well.

When we examine the weighted video topics in the trends, Entertainment, Television Program and Sports are above here, as in other statistical graphics. Football is seen as the most followed sport in the sports category. When we look at the most followed areas of the Music category, which is at the top, Pop, Hip-Hop, Electronic and Asian Music are leading. These three music genres rank high in youtube music trends. In the game category, Action and Adventure type games are very popular. In How-to videos, the Technology category is at the top. This is normal since the needs such as Installation and Promotion are high in Turkish technological product consumers. In addition, the Food Category is at the top.

When we look at the areas where the trending channels produce content, the same topics as the trending videos appear, but here a few different topics show themselves, let’s examine them. Channels producing content on Role-Playing Games and Strategy games have shown themselves in trends, albeit a little. Apart from this, we can say that the video titles they produce are proportional and harmonious.

It cannot be said that the title lengths have a full effect on the statistics of the videos. You can examine it in the correlation table above.

Although the length of the description has a negative correlation with the data of views and likes, there is a positive correlation with the data of comments. This positive correlation tells us that we should be a little more careful and planned in terms of explanations, rather than giving the same template explanations to each video, summarizing the video to the explanation or recognizing the video with key paraphrases will affect the statistics in my opinion. For this, language models that we call Large Language Model can be used. The speech scripts of the videos are extracted, the speech data is extracted and then summarized. For this, it is necessary to create a special dataset and train the models, but it is done.

Above, we have shown the average values of title and description lengths by category. Movies, Educational Content and Music videos have the highest explanation lengths, while the shortest ones are accompanied by videos in the Science and Technology, News and Vehicle categories. The main reason for the length of this short description (although the content is technical) in the Tools and Science category is that trending videos like these are usually Shorts videos and Shorts videos cannot be annotated. Besides, I think this shortness in the News Category is because News channels use a certain template description. The average title lengths of the categories seem to be almost the same with each other, there are only minor differences. For each category, the ideal figure is seen as 8. I want to see if the Titles are related to the Descriptions and the effect of this incidence rate on statistics but let’s not use a language model, I will apply it there when I write a kaggle version of this notebook.

We analyzed Youtube Turkey Trends and we evaluate results with our perspective if your perspective tells different things to you let me know. Thx for reading.