If you train your algorithm with whatever raw data, you would get whatever result. Even a model perfectly analysing the given situation becomes useless when not being adequately trained. In this specific case, the problem is clear: that tool was designed to deal with a different type of scenarios. Coming up with names for objects by training the program with many other names of equivalent objects makes perfect sense. Trying to figure out the best title for an article by analysing a big number of past titles about different subjects makes no sense at all.
The only sensible proceeding in this specific case would have been to rely on a tool able to reasonably analyse article contents and accurately determine the associated title; also to analyse a big amount of contents and output a good summary for them. You train that tool with all the articles during the last years, such that it can come up with the best summary and generate a title from that summary. If they did that, the training might have been considered acceptably good and the accuracy of the used model might have been properly assessed. Under the current conditions, these results don't differ much from the generation of random words.
I think the best way to test this is get a few people who have never read slashdot and see if they can come up with better headlines. I am sure they could, but be interesting to compare.
people who have never read slashdot and see if they can come up with better headlines.
My whole point was that the used methodology is objectively not in a position to deliver a proper understanding of the situation (= to actually summarise what has been happening during the last years in somehow meaningful headlines). It can deliver the most commonly used words or combinations of them, what might be useful for naming a band or a colour but not so much for the title of an article. A better version would have been one able to dismiss incoherent or meaningless sentences. Even people with low te
Thanks for sharing your profounds insights about this article, my comments and even life itself. I am so grateful for having been granted this opportunity to enjoy your wisdom that I am feeling almost like crying. I look forward to your next mind-blowing lesson, professor. LOL.
Problem misunderstanding and bad model training (Score:5, Insightful)
The only sensible proceeding in this specific case would have been to rely on a tool able to reasonably analyse article contents and accurately determine the associated title; also to analyse a big amount of contents and output a good summary for them. You train that tool with all the articles during the last years, such that it can come up with the best summary and generate a title from that summary. If they did that, the training might have been considered acceptably good and the accuracy of the used model might have been properly assessed. Under the current conditions, these results don't differ much from the generation of random words.
Re: (Score:0)
Under the current conditions, these results don't differ much from the generation of random words.
Well, isn't that the essence of Slashdot ?
Re: (Score:2)
Well, isn't that the essence of Slashdot ?
In that case, I correct my statement: this is a perfect result! Excellent work! LOL.
Re: (Score:2)
Re: (Score:2)
people who have never read slashdot and see if they can come up with better headlines.
My whole point was that the used methodology is objectively not in a position to deliver a proper understanding of the situation (= to actually summarise what has been happening during the last years in somehow meaningful headlines). It can deliver the most commonly used words or combinations of them, what might be useful for naming a band or a colour but not so much for the title of an article. A better version would have been one able to dismiss incoherent or meaningless sentences. Even people with low te
Re: (Score:1)
Woosh!
Re: (Score:2)