If interested, this text is written by man.
“Wisconsin is clearly on the winning line, winning 51-10 after the third quarter of the match. The team increased its advantage when Russell Wilson found Jacob Petersen's Paz, and he made a touchdown after a seven-meter tug, making the score 44-3. "
These words begin a fragment of a news note published 60 seconds after the end of the third quarter of a football match between university teams in Wisconsin and Nevada. Although at first glance it is difficult to understand, but this note is completely written by a computer program.
The generator of journalistic texts created in the company Narrative Science , which is developing in the field of artificial intelligence.
The program takes data such as sports match statistics, financial reports of companies or real estate sales data, and turns them into newspaper articles. For years, programmers tried to create such robots journalists for sports reporting, but the result was not brilliant: the texts were dry and it was obvious that the article was generated by a computer. Journalism professor Chris Hammond (Kris Hammond) and computer science professor Larry Birnbaum from the laboratory of intelligent systems at Northwestern University in Illinois have been working in this field for ten years and have managed to achieve a qualitatively better result. So last year appeared a startup Narrative Science.
')
“It seemed to me some kind of magic,” said Roger Lee, a partner at investment company Battery Ventures , which raised $ 6 million in investments for a startup. “It looks like a man wrote it.”
Experts in artificial intelligence systems are also impressed with the success of Narrative Science. In their opinion, Narrative Science demonstrates "an increase in the complexity of automatic recognition and, now, the generation of syntactic structures."
The innovative work of Narrative Science raises a wide range of questions as to whether such AI programs are able to help professional journalists in their work or replace them. High-tech has already influenced the economic reality of the media when online print media revenues have fallen due to online advertising. Will journalist robots become another nail in the coffin of traditional journalism?
The leaders of Narrative Science look at things soberly and position their program solely as a tool for low-budget work under time pressure. At the moment, the company has only 20 customers, and some of them are just experimenting with a strange novelty. Among the clients are traditional media who wish to increase content generation through news notes on the financial results of local companies and more complete coverage of sports matches in youth leagues.
“Thus, we are helping to write articles that otherwise would not have been written at all,” says the director of Narrative Science, answering a question about ousting journalism by computer competitors.
For example, the aforementioned company The Big Ten Network (BTN) began using text generators in the spring of 2010 for short notes on baseball and softball games. They appeared on the site one or two minutes after the end of the game. For the generation of texts, the final match statistics and the online minute-by-minute decoding of the course of the game, which is conducted in real time in all American sports leagues, were used. The program has evolved over time with the help of professional editors at BTN.
Development of Narrative Science can make logical conclusions based on historical data, as well as the sequence and results of previous games. To generate meaningful constructions, the program uses the concepts of “individual contribution”, “team effort”, “strong-willed victory” (gain after lagging behind in the score), “confused advantage”, “record result in the season”, “players on the rise” several matches in a row), “team positions” in different sections of sports statistics. Then the program decides which element is most important in a particular match - and it is selected for the first paragraph of the article. The collected data determine the choice of words. For example, in the case of a large advantage in the account, the program may choose the word “defeat” instead of “victory”.
"The main thing is the composition, the construction of the text," says Chris Hammond. “You can't just take numbers and convert them to words.”
Last fall, BTN expanded its use of the program to football and basketball games as well. This content allows you to attract huge traffic from Google by keywords, because on this site, reports about the game appear earlier than those of competitors, and Google very much appreciates the operational content on hot topics. Last year, traffic to the BTN football section grew by 40% compared with 2009.
Another client of Narrative Science is the publishing and real estate firm Hanley Wood , which with the help of the robot publishes monthly on its website builderonline.com reports on the state of real estate markets in 350 nearby towns and districts. We immediately see where prices are rising or falling, where a change in the volume of transactions is observed. Hiring people to keep track of such trends is too expensive, and information is needed.
This customer worked with Narrative Science for several months to customize the program for a specific data structure, but the result exceeded all expectations. One of the leaders of Hanley Wood, in the past - a professional journalist from the agency Thomson Reuters, admitted that he was amazed at the quality of the articles. “They crossed a big linguistic barrier,” he notes. “Articles are by no means alike.”
Although they have only 20 clients, the company Narrative Science is already earning a good profit. From its customer Hanley Wood, the company charges $ 10 for each article with a volume of 500 words (for comparison, 881 words in this text), that is, about $ 3,500 per month. At the same time, for media sites this is a very good deal - even cheaper than the cost of articles from content factories like Demand Media .
Without a doubt, in the future, the quality of texts of Narrative Science will only improve, and the cost will gradually go down. In the future, anyone can use such programs. According to experts, this opens up new horizons for journalism, since professional reporters get their hands on the most powerful tool for data mining. You can study the topic from all sides and find such relationships that it was impossible to even imagine.
Well, Chris Hammond himself quotes the press with a grin, which writes that in 20 years the computer will be able to get the Pulitzer Prize in journalism. The professor believes that this is not the case. The computer will receive an award not in twenty, but in five years, and the award should be awarded to the authors of the program.
These words begin a fragment of a news note published 60 seconds after the end of the third quarter of a football match between university teams in Wisconsin and Nevada. Although at first glance it is difficult to understand, but this note is completely written by a computer program.
The generator of journalistic texts created in the company Narrative Science , which is developing in the field of artificial intelligence.
The program takes data such as sports match statistics, financial reports of companies or real estate sales data, and turns them into newspaper articles. For years, programmers tried to create such robots journalists for sports reporting, but the result was not brilliant: the texts were dry and it was obvious that the article was generated by a computer. Journalism professor Chris Hammond (Kris Hammond) and computer science professor Larry Birnbaum from the laboratory of intelligent systems at Northwestern University in Illinois have been working in this field for ten years and have managed to achieve a qualitatively better result. So last year appeared a startup Narrative Science.
')
“It seemed to me some kind of magic,” said Roger Lee, a partner at investment company Battery Ventures , which raised $ 6 million in investments for a startup. “It looks like a man wrote it.”
Experts in artificial intelligence systems are also impressed with the success of Narrative Science. In their opinion, Narrative Science demonstrates "an increase in the complexity of automatic recognition and, now, the generation of syntactic structures."
The innovative work of Narrative Science raises a wide range of questions as to whether such AI programs are able to help professional journalists in their work or replace them. High-tech has already influenced the economic reality of the media when online print media revenues have fallen due to online advertising. Will journalist robots become another nail in the coffin of traditional journalism?
The leaders of Narrative Science look at things soberly and position their program solely as a tool for low-budget work under time pressure. At the moment, the company has only 20 customers, and some of them are just experimenting with a strange novelty. Among the clients are traditional media who wish to increase content generation through news notes on the financial results of local companies and more complete coverage of sports matches in youth leagues.
“Thus, we are helping to write articles that otherwise would not have been written at all,” says the director of Narrative Science, answering a question about ousting journalism by computer competitors.
For example, the aforementioned company The Big Ten Network (BTN) began using text generators in the spring of 2010 for short notes on baseball and softball games. They appeared on the site one or two minutes after the end of the game. For the generation of texts, the final match statistics and the online minute-by-minute decoding of the course of the game, which is conducted in real time in all American sports leagues, were used. The program has evolved over time with the help of professional editors at BTN.
Development of Narrative Science can make logical conclusions based on historical data, as well as the sequence and results of previous games. To generate meaningful constructions, the program uses the concepts of “individual contribution”, “team effort”, “strong-willed victory” (gain after lagging behind in the score), “confused advantage”, “record result in the season”, “players on the rise” several matches in a row), “team positions” in different sections of sports statistics. Then the program decides which element is most important in a particular match - and it is selected for the first paragraph of the article. The collected data determine the choice of words. For example, in the case of a large advantage in the account, the program may choose the word “defeat” instead of “victory”.
"The main thing is the composition, the construction of the text," says Chris Hammond. “You can't just take numbers and convert them to words.”
Last fall, BTN expanded its use of the program to football and basketball games as well. This content allows you to attract huge traffic from Google by keywords, because on this site, reports about the game appear earlier than those of competitors, and Google very much appreciates the operational content on hot topics. Last year, traffic to the BTN football section grew by 40% compared with 2009.
Another client of Narrative Science is the publishing and real estate firm Hanley Wood , which with the help of the robot publishes monthly on its website builderonline.com reports on the state of real estate markets in 350 nearby towns and districts. We immediately see where prices are rising or falling, where a change in the volume of transactions is observed. Hiring people to keep track of such trends is too expensive, and information is needed.
This customer worked with Narrative Science for several months to customize the program for a specific data structure, but the result exceeded all expectations. One of the leaders of Hanley Wood, in the past - a professional journalist from the agency Thomson Reuters, admitted that he was amazed at the quality of the articles. “They crossed a big linguistic barrier,” he notes. “Articles are by no means alike.”
Although they have only 20 clients, the company Narrative Science is already earning a good profit. From its customer Hanley Wood, the company charges $ 10 for each article with a volume of 500 words (for comparison, 881 words in this text), that is, about $ 3,500 per month. At the same time, for media sites this is a very good deal - even cheaper than the cost of articles from content factories like Demand Media .
Without a doubt, in the future, the quality of texts of Narrative Science will only improve, and the cost will gradually go down. In the future, anyone can use such programs. According to experts, this opens up new horizons for journalism, since professional reporters get their hands on the most powerful tool for data mining. You can study the topic from all sides and find such relationships that it was impossible to even imagine.
Well, Chris Hammond himself quotes the press with a grin, which writes that in 20 years the computer will be able to get the Pulitzer Prize in journalism. The professor believes that this is not the case. The computer will receive an award not in twenty, but in five years, and the award should be awarded to the authors of the program.
Source: https://habr.com/ru/post/128281/
All Articles