This section talks about the entries of a sentiment model.
Entries are the elements that define units that are going to have a specific sentiment behavior. With a new sentiment model you can add sentiment behavior to words or groups of words to improve the results of your application in specific domains.
When you access a sentiment model, you will be able to add new entries just by introducing their
form/lemma and their sentiment behavior. The sentiment behavior will be determined by two fields:
entry type and
entry type, defines what type of sentiment behavior the entry is going to have. There are three possible behaviors to choose from:
It's also possible to leave this field blank for the cases where the entry by itself does not have a behavior associated.
The most common value will be POLARITY, as the most common negators and modifiers are already taken into account in the General model we provide. The only cases where you will define an entry that does not have an
entry type value associated are:
For instance, in English, "good" is a word with a positive polarity, "not" is the most common negator, and "very" is a modifier that increases the strenght of the word with polarity that it accompanies. You'd left the
entry type blank to define that "sentence" does not have any sentiment behavior associated, except when you talk about a "prison sentence", which can be considered negative and would be defined as a subentry.
The second field,
value, selects the value associated to entries of the types POLARITY and MODIFIER.
For entries of the POLARITY type, there are the following seven values:
For entries of the MODIFIER type, there are have the following four values:
This is what you will see when you add a new entry:
If one of the values specified is not correct, you will see an error message specifying why the entry cannot be created. There are two types of errors:
form which has to be unique and is limited to 255 characters.
For each entry created you will see its basic information: its form/lemma, entry type, value (in the cases where it applies), the number of subentries associated and when it was last edited.
In the first column of the table there are two actions associated to the entry:
Every entry created will be shown in this table, where you will be able to select how many entries to show in each page, order them by any of the columns and filter them dynamically by text appearance. This dynamic filtering will also include the
aliases field, even though it is not shown in the summary.
To modify any field of an entry, you will have to access its editing view. From there, you can modify all of its information.
Both the sidebar and the main panel feature a question mark, , which shows a small tour that explains briefly each field.
The fields are divided in two sections:
The basic fields of an entry are
Entry Type and
Value (in the
entry type values it applies). Only
Entry type and the
value(when it applies) are mandatory when creating an entry.
Form/Lemma: is what defines the entry. It's mandatory and it's the word or group of words where the sentiment information you define will be taken into account in an analysis. Any words specified here, will be considered a lemma first if they coincide with one, otherwise, they will be considered a form.
The difference between a lemma and a form is that a lemma groups several words (or forms). For instance, "be" would group the forms "be", "is", "are", "was", "were" and any other tense of the verb "be". If you were to include as an entry the words "be nice", then you would be considering "be nice", "is nice", "are nice", "was nice", etc. On the other hand, if the entry defined is "is nice", then that's the only form it will detect.
form/lemma of the entry. Each aliases must go in a new line, and they are interpreted like the
form/lemma: if the word can be a lemma, it will be considered as such, otherwise, it will be considered a lexical form.