Krovi: Big Data - as dream. 8th series. Non-technical. Modular journalism
In previous episodes: Big Data is not just a lot of data. Big Data is a process with positive feedback. "The Obama button" as the embodiment of rtBD & A. Philosophy of the development of Big Data. BD - also Bolt Data. BD analysts. In this series, let's talk about the impact of BD on such a non-technical industry as journalism.
Very schematically - programmers are akin to journalists: both industries use past skills (they use knowledge of previous generations, developers of methods and languages), both professions are aimed at modernizing what is happening and what is happening (to improve the lives of specific people), leading people in both directions seek to rise to the next level aimed at the future is to become architects and writers who influence huge teams of followers and nations.
Both professions have changed a lot over the past decades: stenographers and typists disappear together, proofreaders disappear, the interim âoperator PCâ disappears into the past, and Visual Basic and the Internet have repeatedly expanded the âfront doorâ to both professions. Creation of programs and the publication of their own opinions became available "and old and young."
And both professions go under the sword of Damocles for templating: the rate at which new tasks and information flows from all corners, which have become very small, of the earth globe, flowed, automatically led to an increase in replicable cliches. More and more programs are being âcompiledâ from modules-libraries, more and more journalistic short news materials can be collected automatically from typical strapping around âbare factsâ. Experiments on programming automation and writing articles by artificial intelligence have already entered the preindustrial phase.
')
But there should be no pessimism: mankind is not scattered by top-notch experts - the usual transfer from drying out industries to gaining strength will occur. About 15 years ago, when the first Internet wave in journalism had already formed, at the meeting in the office of Yassen Zassoursky, dean of the Faculty of Journalism of Moscow State University, the question â Well, is there really a lot of journalists here? !! â Yasen Nikolayevich sadly shook his head. In the large office there were several tables, on which Mount Everest books, magazines, newspapers, almanacs, saturated with thoughts and time, stood on the heights of the mountains. And we had a hot hour discussing with the doctor of philology where to turn the path of future graduates of journalism. Honestly, life still turned out to be richer than our speculations: blogs and Facebook, YouTube and news aggregators, peculiar to the uniqueness of Twitter, communities around traveling photographers and like-minded commentators around all sorts of thematic resources - all this diversity could not even appear in most pink dreams at the turn of the century. It is symptomatic that the grandson of Yasen Nikilaevich, âInternet researcherâ Ivan Zassoursky, heads the New Media department at the Faculty of Journalism.
We will continue the positive dose and see which trends can strengthen the position of leading journalists in their desire to influence all large groups of people, and how the Big Data industry can help with this re / evolutionary explosion:
1. Identify new trends
A typical example is the Chelyabinsk meteorite. Those. we are talking about a completely new entity that previously did not exist, or an object whose information background was minimal for a long time.
2. Change in existing trends
For example, elections in France: there are three political forces and three persons (Hollande, Sarkozy, Marie Le Pen), there is a division of society into those who have already made their choice and those who hesitate. The goal of the âplayersâ is to get the largest possible percentage of the votes of the undecided voters. You should constantly analyze trends and reactions to actions and campaign promises.
3. Ratings of interests (media persons, thematic events)
In any field of society there are leaders recognizable by society: these are individuals (football players, politicians, scientists, musicians), and objects (Sosa-Cola, Moscow, Sberbank, Zenit). The very fact of getting into a thematic rating or a change in positions in ratings is a signal of changes in the public information field.
In the previous series, it was already said about the need to distinguish between approaches:
A) personalized multi-data-per-object,
B) the information field of data in the industry and around objects.
Until recently, for journalists and the public, which received the main stream of data almost exclusively from the media, there was in fact no alternative to option âAâ (topic âDossierâ). Newcomers (people, brands) needed to actively interact with the press in order to get into the circle of objects mentioned by the media.
With the development of social media, the direction of the vector has changed significantly:
- âSensorsâ (people, companies, structures, robots) generate and constantly generate data in the information field;
- The media adapts to the new reality - because of the competition for the speed of news release, the editorial staff creates monitoring groups for the early detection of information leads;
- The role of the media is changing - because of the sharply increased overall flow of materials in the world, thematic focusing and polarization of the opinions of specific media takes place;
- The role of the media is declining - people are increasingly using other information flows, competition is becoming tougher over the time of people, which is 24 hours a day.
Here are the interesting data obtained from Brand Analytics on the Russian-speaking information field for the week of April:
Date - News - News Comments
04/02/2015 ..... 147 607 ... 68 957
04/03/2015 ..... 126 685 ..103 503
04.04.2015 ... 69 924 ... 85 015
04/05/2015 ... 58,961 ... 78,819
04/06/2015 ... 121,247 ... 92,784
04/07/2015 ... 148 011 ... 104 245
04/08/2015 ... 189 650 ... 92 011
Those. 862 thousand news was published in a week , to which 625 thousand comments were made .
For those who are more comfortable to think in the data format for the month - data for March: news - 3.7 million, comments to them - 2.7 million. In total, 6.4 million âabout-media articlesâ are received. Quite a bit (0.5%) of the flow in 1 BILLION of Russian-language messages in social networks (plus a few blogs, forums, review sites, etc.) generated by people who are âsensorsâ around the world.
Therefore, it is not surprising that an unexpectedly occurring phenomenon - the same âChelyabinsk meteoriteâ - took just a few minutes to appear in social networks (Twitter, YouTube, VKontakte) and as much as 2-3 hours for the first information to appear in the media.
Thus, modern media (no matter what format they currently operate) make it very important to quickly receive information from the âsensorsâ (manufacturers' sites, people at the scene, robotic systems), thematic filtering to reduce the âomnivorousnessâ and more targeted readers, as well as operational rating , which allows you to âbe in good shapeâ and in time to adjust your view of the world on how the world is perceived by ordinary people.
As a real case, let's look at three ratings that are regularly produced by the largest Russian media based on the analysis of large volumes of unstructured information from social media:
1. Profile, thematic: monthly citation rating of the Russian media on Lenta.ru
lenta.ru/articles/2015/04/07/bamarch
2. Non-core, thematic, situational: during the World Cup in Brazil, the site RT (Russia Today) published a daily rating of the most popular âhumanityâ materials on the football theme
br-analytics.ru/blog/sovmestnyj-proekt-brand-analytics-i-russia-today-ezhednevny
3. Profile, total: monthly rating of media persons on the RIA Novosti website:
ria.ru/infografika/20150402/1055894152.html
Regarding the last rating - media people - of course, much more interesting is the daily rating, reflecting changes in the perception of society. Here are the Top 50 tables for yesterday (April 26): a separate media rating and a social media rating. Find many differences :-)
The "+" sign next to the person means that this media person was included in today's Top 50 rating for the day and was absent from the Top 50 yesterday. This rating allows you to immediately assess how much you are aware of yesterday's events of the day - do you understand what exactly served to make Nazarbayev appear on the list (hint: yesterday the presidential elections were held in Kazakhstan).
Media rating:
No. Quantity Fiz. face
1 4528 Vladimir Putin
2 1435 Petro Poroshenko
3 1070 Barack Obama
4,959 Nursultan Nazarbayev
5 692 Wladimir Klitschko
6,590 Ramzan Kadyrov
7,438 Arseniy Yatsenyuk
8 390 Joseph Stalin
9 387 Dmitry Medvedev
10 375 Dmitry Rogozin
11 348 Dmitry Peskov
12 337 Adolf Hitler
13 310 Angela Merkel
14,296 Boris Nemtsov
15 295 Vitali Klitschko
16 257 Vladimir Lenin
17,252 Francois Hollande
18 243 Viktor Yanukovych
19 240 Boris Yeltsin
20 209 Alexander Lukashenko
21 188 Rinat Akhmetov
22 188 Joseph Kobzon
23 177 Sergey Shoigu
24,152 Federica Mogherini +
25,115 Ban Ki-moon
26 145 Sergey Lavrov
27 137 Vladimir Soloviev +
28 131 Stepan Bandera
29 123 Yulia Tymoshenko
30 119 Mikhail Saakashvili
31 118 Mikhail Khodorkovsky +
32 115 German Gref +
33 111 Alexander Zakharchenko
34 103 Jose Mourinho
35 98 Robert Downey +
36 96 Kim Kardashian +
37 90 Alexey Navalny
38 86 Mikhail Gorbachev +
39 82 Fabio Capello
40 77 Nadezhda Savchenko
41 75 Alexander Pushkin
42 73 John Kerry
43 72 Oleg Lyashko
44 68 Lionel Messi
45 68 Napoleon Bonaparte +
46 65 Igor Kolomoisky
47 65 Nikita Mikhalkov +
48 65 Jen Psaki +
49 64 Arkady Rotenberg +
50 59 Cristiano Ronaldo +
Rating of Social Media:
No. Quantity Fiz. face
1 144539 Vladimir Putin
2 35769 Petro Poroshenko
3 31902 Barack Obama
4 25706 Ramzan Kadyrov
5 23137 Joseph Stalin
6 18729 Vitali Klitschko
7 14936 Adolf Hitler
8 14328 Arseniy Yatsenyuk
9 13440 Wladimir Klitschko +
10 13027 Alexander Pushkin
11 12799 Vladimir Lenin
12 12267 Nursultan Nazarbayev +
13 11995 harry potter
14 11579 Boris Nemtsov
15 9377 Cristiano Ronaldo
16 9326 Erich Maria Remarque
17 8816 Lev Tolstoy
18 8469 Timati
19,8405 Lionel Messi
20 7974 Albert Einstein
21 7799 William Shakespeare
22 7606 Omar Khayam
23 7429 Sergey Yesenin
24 7156 Viktor Yanukovych
25 6843 Sergey Shoigu
26 6452 Dmitry Rogozin
27 6402 Vysotsky Vladimir
28 6189 Dmitry Medvedev
29 5974 Bernard Shaw +
30 5838 Vera Brezhnev
31 5661 Vladimir Solovyov +
32 5642 Alexey Navalny
33 5622 Yulia Tymoshenko
34 5608 Dmitry Peskov
35 5526 Vladimir Zhirinovsky
36 5513 Fabio Capello
37 5426 Zemfira +
38 5391 Johnny Depp
39 5198 Napoleon Bonaparte
40 5006 Anatoly Shary
41 4943 Stepan Bandera
42 4934 Andrey Lenitsky +
43 4929 Boris Yeltsin +
44 4922 Fyodor Dostoevsky
45 4899 Sergey Lavrov
46 4583 Joseph Brodsky +
47 4572 Faina Ranevskaya +
48 4467 Nietzsche
49 4467 Polina Gagarina +
50 4446 Stephen King +
For the formation of a daily rating, a stream of 30-40 million Russian-language messages is processed, taking into account ânon-object linguistics of social mediaâ , which allows identifying new entities, and not just filtering data with a predefined list of people.
Let us summarize and continue our conversation with Yasen Nikolayevich on the topic âHow can more readers please the forces of a small number of journalists?â As a technocratic approach, we offer âmodular journalismâ:
- The information sea generated by âsensorsâ (people, companies, structures, robots) - for the Russian-speaking segment it is 30-50 million messages per day from 12-15 million âsensorsâ;
- Analytical system that reveals new trends (âChelyabinsk meteoriteâ, âEarthquake in Nepalâ);
- Analytical system that tracks changes in existing trends ;
- Analytical system, highlighting named entities (objects) and ranking media objects;
- Rating system for monitoring thematically-related objects.
Such a set of modules will allow you to generate from 10 to 90% of materials on your content resource, allowing you to quickly respond to expectations in the information support of readers. In one way or another, most of the publications have long been moving in this direction (for example, a selection of 5 news on the popular TJournal) - itâs all about the tools used and the âtrustâ in us, readers, in terms of understanding our interests.
Of course, an intelligent reader will immediately see the danger of template making and âkillingâ publications - if every publication (even if it is within its scope, there are still a lot of such) will publish materials on the principle of âmodular journalismâ, then why are there so many similar publications! We do not have an answer to this question now, but we are sure that creative people will definitely find a creative way to the next level :-)
UPD: Popularity ratings on the commentary of News Sites, blogs LJ, Twitter, YouTube, VK are available in the public domain on pl.ilook.ru
PS For those bloggers, journalists, editors who need or are interested to be in the know, you can send a request to pr@br-anaytics.ru with a request to include your address in the daily mailing list of Media Persons for the past day. Our colleagues from Brand Analytics are ready to provide free information with ratings of TOP-50 media people until the end of May.
Very schematically - programmers are akin to journalists: both industries use past skills (they use knowledge of previous generations, developers of methods and languages), both professions are aimed at modernizing what is happening and what is happening (to improve the lives of specific people), leading people in both directions seek to rise to the next level aimed at the future is to become architects and writers who influence huge teams of followers and nations.
Both professions have changed a lot over the past decades: stenographers and typists disappear together, proofreaders disappear, the interim âoperator PCâ disappears into the past, and Visual Basic and the Internet have repeatedly expanded the âfront doorâ to both professions. Creation of programs and the publication of their own opinions became available "and old and young."
And both professions go under the sword of Damocles for templating: the rate at which new tasks and information flows from all corners, which have become very small, of the earth globe, flowed, automatically led to an increase in replicable cliches. More and more programs are being âcompiledâ from modules-libraries, more and more journalistic short news materials can be collected automatically from typical strapping around âbare factsâ. Experiments on programming automation and writing articles by artificial intelligence have already entered the preindustrial phase.
')
But there should be no pessimism: mankind is not scattered by top-notch experts - the usual transfer from drying out industries to gaining strength will occur. About 15 years ago, when the first Internet wave in journalism had already formed, at the meeting in the office of Yassen Zassoursky, dean of the Faculty of Journalism of Moscow State University, the question â Well, is there really a lot of journalists here? !! â Yasen Nikolayevich sadly shook his head. In the large office there were several tables, on which Mount Everest books, magazines, newspapers, almanacs, saturated with thoughts and time, stood on the heights of the mountains. And we had a hot hour discussing with the doctor of philology where to turn the path of future graduates of journalism. Honestly, life still turned out to be richer than our speculations: blogs and Facebook, YouTube and news aggregators, peculiar to the uniqueness of Twitter, communities around traveling photographers and like-minded commentators around all sorts of thematic resources - all this diversity could not even appear in most pink dreams at the turn of the century. It is symptomatic that the grandson of Yasen Nikilaevich, âInternet researcherâ Ivan Zassoursky, heads the New Media department at the Faculty of Journalism.
We will continue the positive dose and see which trends can strengthen the position of leading journalists in their desire to influence all large groups of people, and how the Big Data industry can help with this re / evolutionary explosion:
1. Identify new trends
A typical example is the Chelyabinsk meteorite. Those. we are talking about a completely new entity that previously did not exist, or an object whose information background was minimal for a long time.
2. Change in existing trends
For example, elections in France: there are three political forces and three persons (Hollande, Sarkozy, Marie Le Pen), there is a division of society into those who have already made their choice and those who hesitate. The goal of the âplayersâ is to get the largest possible percentage of the votes of the undecided voters. You should constantly analyze trends and reactions to actions and campaign promises.
3. Ratings of interests (media persons, thematic events)
In any field of society there are leaders recognizable by society: these are individuals (football players, politicians, scientists, musicians), and objects (Sosa-Cola, Moscow, Sberbank, Zenit). The very fact of getting into a thematic rating or a change in positions in ratings is a signal of changes in the public information field.
In the previous series, it was already said about the need to distinguish between approaches:
A) personalized multi-data-per-object,
B) the information field of data in the industry and around objects.
Until recently, for journalists and the public, which received the main stream of data almost exclusively from the media, there was in fact no alternative to option âAâ (topic âDossierâ). Newcomers (people, brands) needed to actively interact with the press in order to get into the circle of objects mentioned by the media.
With the development of social media, the direction of the vector has changed significantly:
- âSensorsâ (people, companies, structures, robots) generate and constantly generate data in the information field;
- The media adapts to the new reality - because of the competition for the speed of news release, the editorial staff creates monitoring groups for the early detection of information leads;
- The role of the media is changing - because of the sharply increased overall flow of materials in the world, thematic focusing and polarization of the opinions of specific media takes place;
- The role of the media is declining - people are increasingly using other information flows, competition is becoming tougher over the time of people, which is 24 hours a day.
Here are the interesting data obtained from Brand Analytics on the Russian-speaking information field for the week of April:
Date - News - News Comments
04/02/2015 ..... 147 607 ... 68 957
04/03/2015 ..... 126 685 ..103 503
04.04.2015 ... 69 924 ... 85 015
04/05/2015 ... 58,961 ... 78,819
04/06/2015 ... 121,247 ... 92,784
04/07/2015 ... 148 011 ... 104 245
04/08/2015 ... 189 650 ... 92 011
Those. 862 thousand news was published in a week , to which 625 thousand comments were made .
For those who are more comfortable to think in the data format for the month - data for March: news - 3.7 million, comments to them - 2.7 million. In total, 6.4 million âabout-media articlesâ are received. Quite a bit (0.5%) of the flow in 1 BILLION of Russian-language messages in social networks (plus a few blogs, forums, review sites, etc.) generated by people who are âsensorsâ around the world.
Therefore, it is not surprising that an unexpectedly occurring phenomenon - the same âChelyabinsk meteoriteâ - took just a few minutes to appear in social networks (Twitter, YouTube, VKontakte) and as much as 2-3 hours for the first information to appear in the media.
Thus, modern media (no matter what format they currently operate) make it very important to quickly receive information from the âsensorsâ (manufacturers' sites, people at the scene, robotic systems), thematic filtering to reduce the âomnivorousnessâ and more targeted readers, as well as operational rating , which allows you to âbe in good shapeâ and in time to adjust your view of the world on how the world is perceived by ordinary people.
As a real case, let's look at three ratings that are regularly produced by the largest Russian media based on the analysis of large volumes of unstructured information from social media:
1. Profile, thematic: monthly citation rating of the Russian media on Lenta.ru
lenta.ru/articles/2015/04/07/bamarch
2. Non-core, thematic, situational: during the World Cup in Brazil, the site RT (Russia Today) published a daily rating of the most popular âhumanityâ materials on the football theme
br-analytics.ru/blog/sovmestnyj-proekt-brand-analytics-i-russia-today-ezhednevny
3. Profile, total: monthly rating of media persons on the RIA Novosti website:
ria.ru/infografika/20150402/1055894152.html
Regarding the last rating - media people - of course, much more interesting is the daily rating, reflecting changes in the perception of society. Here are the Top 50 tables for yesterday (April 26): a separate media rating and a social media rating. Find many differences :-)
The "+" sign next to the person means that this media person was included in today's Top 50 rating for the day and was absent from the Top 50 yesterday. This rating allows you to immediately assess how much you are aware of yesterday's events of the day - do you understand what exactly served to make Nazarbayev appear on the list (hint: yesterday the presidential elections were held in Kazakhstan).
Media rating:
No. Quantity Fiz. face
1 4528 Vladimir Putin
2 1435 Petro Poroshenko
3 1070 Barack Obama
4,959 Nursultan Nazarbayev
5 692 Wladimir Klitschko
6,590 Ramzan Kadyrov
7,438 Arseniy Yatsenyuk
8 390 Joseph Stalin
9 387 Dmitry Medvedev
10 375 Dmitry Rogozin
11 348 Dmitry Peskov
12 337 Adolf Hitler
13 310 Angela Merkel
14,296 Boris Nemtsov
15 295 Vitali Klitschko
16 257 Vladimir Lenin
17,252 Francois Hollande
18 243 Viktor Yanukovych
19 240 Boris Yeltsin
20 209 Alexander Lukashenko
21 188 Rinat Akhmetov
22 188 Joseph Kobzon
23 177 Sergey Shoigu
24,152 Federica Mogherini +
25,115 Ban Ki-moon
26 145 Sergey Lavrov
27 137 Vladimir Soloviev +
28 131 Stepan Bandera
29 123 Yulia Tymoshenko
30 119 Mikhail Saakashvili
31 118 Mikhail Khodorkovsky +
32 115 German Gref +
33 111 Alexander Zakharchenko
34 103 Jose Mourinho
35 98 Robert Downey +
36 96 Kim Kardashian +
37 90 Alexey Navalny
38 86 Mikhail Gorbachev +
39 82 Fabio Capello
40 77 Nadezhda Savchenko
41 75 Alexander Pushkin
42 73 John Kerry
43 72 Oleg Lyashko
44 68 Lionel Messi
45 68 Napoleon Bonaparte +
46 65 Igor Kolomoisky
47 65 Nikita Mikhalkov +
48 65 Jen Psaki +
49 64 Arkady Rotenberg +
50 59 Cristiano Ronaldo +
Rating of Social Media:
No. Quantity Fiz. face
1 144539 Vladimir Putin
2 35769 Petro Poroshenko
3 31902 Barack Obama
4 25706 Ramzan Kadyrov
5 23137 Joseph Stalin
6 18729 Vitali Klitschko
7 14936 Adolf Hitler
8 14328 Arseniy Yatsenyuk
9 13440 Wladimir Klitschko +
10 13027 Alexander Pushkin
11 12799 Vladimir Lenin
12 12267 Nursultan Nazarbayev +
13 11995 harry potter
14 11579 Boris Nemtsov
15 9377 Cristiano Ronaldo
16 9326 Erich Maria Remarque
17 8816 Lev Tolstoy
18 8469 Timati
19,8405 Lionel Messi
20 7974 Albert Einstein
21 7799 William Shakespeare
22 7606 Omar Khayam
23 7429 Sergey Yesenin
24 7156 Viktor Yanukovych
25 6843 Sergey Shoigu
26 6452 Dmitry Rogozin
27 6402 Vysotsky Vladimir
28 6189 Dmitry Medvedev
29 5974 Bernard Shaw +
30 5838 Vera Brezhnev
31 5661 Vladimir Solovyov +
32 5642 Alexey Navalny
33 5622 Yulia Tymoshenko
34 5608 Dmitry Peskov
35 5526 Vladimir Zhirinovsky
36 5513 Fabio Capello
37 5426 Zemfira +
38 5391 Johnny Depp
39 5198 Napoleon Bonaparte
40 5006 Anatoly Shary
41 4943 Stepan Bandera
42 4934 Andrey Lenitsky +
43 4929 Boris Yeltsin +
44 4922 Fyodor Dostoevsky
45 4899 Sergey Lavrov
46 4583 Joseph Brodsky +
47 4572 Faina Ranevskaya +
48 4467 Nietzsche
49 4467 Polina Gagarina +
50 4446 Stephen King +
For the formation of a daily rating, a stream of 30-40 million Russian-language messages is processed, taking into account ânon-object linguistics of social mediaâ , which allows identifying new entities, and not just filtering data with a predefined list of people.
Let us summarize and continue our conversation with Yasen Nikolayevich on the topic âHow can more readers please the forces of a small number of journalists?â As a technocratic approach, we offer âmodular journalismâ:
- The information sea generated by âsensorsâ (people, companies, structures, robots) - for the Russian-speaking segment it is 30-50 million messages per day from 12-15 million âsensorsâ;
- Analytical system that reveals new trends (âChelyabinsk meteoriteâ, âEarthquake in Nepalâ);
- Analytical system that tracks changes in existing trends ;
- Analytical system, highlighting named entities (objects) and ranking media objects;
- Rating system for monitoring thematically-related objects.
Such a set of modules will allow you to generate from 10 to 90% of materials on your content resource, allowing you to quickly respond to expectations in the information support of readers. In one way or another, most of the publications have long been moving in this direction (for example, a selection of 5 news on the popular TJournal) - itâs all about the tools used and the âtrustâ in us, readers, in terms of understanding our interests.
Of course, an intelligent reader will immediately see the danger of template making and âkillingâ publications - if every publication (even if it is within its scope, there are still a lot of such) will publish materials on the principle of âmodular journalismâ, then why are there so many similar publications! We do not have an answer to this question now, but we are sure that creative people will definitely find a creative way to the next level :-)
UPD: Popularity ratings on the commentary of News Sites, blogs LJ, Twitter, YouTube, VK are available in the public domain on pl.ilook.ru
PS For those bloggers, journalists, editors who need or are interested to be in the know, you can send a request to pr@br-anaytics.ru with a request to include your address in the daily mailing list of Media Persons for the past day. Our colleagues from Brand Analytics are ready to provide free information with ratings of TOP-50 media people until the end of May.
Source: https://habr.com/ru/post/256775/
All Articles