In the modern world, the main source of information about household goods for the buyer is the Internet, through which you can easily and quickly find, compare, select and even buy a model of a product of one type or another. However, few buyers think about what is behind the formation of the content on the basis of which they make this choice.
Nevertheless, both the store owner and his marketing department are well aware of the need to provide a high-quality, comprehensive and at the same time fairly easy to understand description for the goods for the goods.
Content about the product can conditionally be divided into several types:
- General information (photo, video, editorial text).
- SEO (body kit for promotion, search, snippets, micro-layout).
- Technical description (characteristics).
- Documentation files.
- Related products (related, dependent, mandatory, alternative, etc.).
Information model
In this article we will look at the third type of content - technical description (in fact, almost everything else can be obtained from this content).
Such content, as a rule, is a table consisting of “Attribute: Value” pairs. At the same time, an attribute should be understood as a separate property of a group of closely related products, reflecting their functional, physical and / or consumer qualities. The value of an attribute is the implementation of such a property in this particular product (in algebraic language, the value is the action of the attribute on the product).
Attributes can be numeric, categorical, and text.
, . , (, , , ..) (), ( ). . ( ).
(, , , ..) (, ) .
, , .. , ( ).
— . , , . , ( , ), “ ”, . , ( — ) .
, , , “ ” , , , , , .
: , , ( ) , , , , - -.
, - . , ( : ) , ( : , .. ).
(, , ) . , , . , .
( : ) (, ).
, (), , — , , . , .. , .
, , ( ), , . , , . , , .
, , .
- . -, API , , . , , .
, -, -. , , .
, :
, .
, — , , , — . ( ) . .
:
- .
- ( ).
- () .
- / .
- .
- ( ).
- .
-, , () .
- , ?
- “”?
- ?
( , , ), , , . , , ( ) , (.. ).
, (), , ( ).
. , A, B, C, 5 ( n=3) ({A,B,C}, {A+B,C}, {A+C,B}, {A,B+C}, {A+B+C}). , .
, . A , , , . , , .
“” . -, . , , - . , . , , . , , .. . , , .
, - .
. , , . , . . , , (, ).
?
, . .
, 10 . , — , . 1000 10 , 100 , .
, , . , ( ) . , “”.
/, . , , ( ) . - . - .
, , , / , , . .
, , ? , , .
. -, . , — . , , 1-2% , 100% .
, , , , .
Data Science , , . , ().
-, — , , / . , . , , .
-, “”, - . .
, , , , ..
, , . , 5-10%, /, .
, “/”-. — ( ) {‘’,’’}. , “/”-. . , , “/”- ‘’, — ‘’.
. , . , /, “/”- .
, , , .
“/”- , ( “---”), “ ” , , /, ( ).
“/”- — - , . : “/”-, , .
, , , . .
, , , ( ) . , , , , .
, , . , .
, .
, -, . , , , .
.
. , “” .
, / . , , , , .
, . / .
, , , , , . , SEO- , .
, — . . , . , , , .
, , , — , , , ..
, :
- Everyone has their own information model, therefore, for adequate interaction of information models, you need to create an internal receiving model.
- The internal model should be atomized, analytical.
- More numeric attributes, less textual!
- The model should correspond to the group of similar products so that the difference between the goods of the group in the number of attributes used is no more than 5-10%.
- Product groups should not be small.
- The language (human) of the internal model and the connected trusted sources must be English, and the system of measures - SI.
- Use the Yes / No model with caution, but it is better not to use it at all.