How do the publications grouped in stories?

1. Texts are compared by the system not by the exact and full match of the texts, but by the presence of words that are defined by the system as keywords for these texts. These key words are determined by the system itself, they do not correspond to the words for highlighting in the topic/topics.
2. The document that entered the system before all other documents in the story is defined as the main publications of stories. Usually, this is the document that appeared earlier by date and time in the system, that is, it is not necessarily the first publication in the media, because different sources are scanned by the system at different speeds.
3. Stories generation cannot translate keywords from one language to another, so usually similar documents in different languages are not combined into one story. Documents in different languages can be combined into one story only if keywords have the same spelling in different languages.
4. Stories generationdoes not consider document tone of voice, autotags, tags and other document parameters, so publications with different tone of voice can be included in the same story. Users with rights can change the toneof voice of the entire story. Learn more about changing the tone of publications in the Story View mode.
   

Creating Stories

You can create stories:

1. in one selected topic;
2. in two or more selected topics;
3. in all topics;
4. by selecting one or more tags.
   

How to select one, several, or all topics, see here.

In the Publications Actions panel, click the stories generation icon. The currently filtered publications will automatically be grouped into stories. After a while, a message about generating stories appears, click the "Confirm" button.

Displaying stories

Stories are displayed as feeds of main publications of stories , displaying only the features of that main publications of stories.

The publications action bar on the left shows the number of documents and the number of generated stories.

The action bar contains a window for sorting the generated stories: by the number of documents in the story, by the total coverage of the stories, by the total Media Visibility Index of the story, by date - from new to old or vice versa.

Please note:

Sorting by date is based on the date of the main publication of the story, which is automatically determined by the system as the beginning of the story. In some cases, there may be errors in the time of publication, depending on the specifics of the source, so in order to find the first publication on a particular topic, it is advisable to check the correctness of the story and the selection of the first document. 

The Media visibility index of all documents is displayed on the documents forming the story (Learn more about Media visibility index).

Story documents display the total coverage of all documents in the story (How Semantrum calculates the forecusted number of views of the publication).

Viewing stories

To view the publications in a story, click View Publications Menu  , the number indicates the number of publications included in the story. This number does not include the main publication of stories. That is, if the number 38 is near the menu of documents in the story, the story has a total of 39 documents (main publication of stories + 38 similar documents). If the story consists only of the main publication of stories (i.e. the system has not found any similar documents), then there will be a 0 next to the menu of documents in the story. When the user clicks on a story-forming document, the user will be taken to that document, not to the story.

Please note:

1. The story feature can handle up to 3,000 publications. If you have more publications filtered when creating stories, the system will only plot the first 3,000 publications. Other documents will not be counted and will not appear in the story list. The more documents in the sample, the longer the stories generation takes.
2. There is no need to select the publications for stories generation.
3. The smaller the array, the more accurate the stories generation. The main publications of stories for these large stories may not be the most representative document, so you should always look inside the story when analyzing stories in large arrays.