"Recommender System"--based on knowledge recommendation

"Recommender System an Introduction". The fourth chapter, based on the knowledge recommendation.

Profile

To acquire and maintain this knowledge at a relatively small cost.

However, in some scenarios, such as housing, automobiles, computers and other goods, the synergy system will be poor due to the lack of scoring data, or time-span factors are very important situation, user preferences with various situations change, etc. these scenarios are not collaborative filtering and content-based recommendations are good at solving.

A knowledge-based referral system solves these problems, and it does not need to score data. Therefore, there is no startup problem.

Based on the knowledge recommendation, the interaction is very strong, so it is a conversational system.

Based on knowledge recommendation is not only a filtering system, but more broadly "in a personalized way to guide users to find interesting or useful items in a large number of potential candidates." or produce these items as output "of the system.

The two basic types of knowledge-based recommender systems are: constraint-based recommendation and case-based recommendation.

Their referral process is similar: the user must specify the requirements, and the system tries to give the solution. Assuming no workaround is found, the user must change the requirements. In addition, the system should also give the explanation of the recommended items.

They differ in how to use the knowledge provided. The case-based recommendation system focuses on retrieving similar items based on different similarity measurements. A constraint-based recommendation system relies on a set of recommended rules that are clearly defined.

Knowledge representation and Reasoning

Knowledge-based systems rely on specific knowledge of item characteristics.

Constraint-based recommendation issues can generally be expressed as constraints that are addressed by the constraint Solver, or as a form of a pull query that is run and resolved through the database engine.

The case-based recommendation system mainly uses the similarity measure standard to retrieve items from the folder.

Constraints

Constraint satisfaction problem (CSP), based on the CSP algorithm and the recommended knowledge base, can be used to construct a constrained recommendation system.

Constraint-based recommender systems involve variables and constraints such as the following:

(1) User attributes (Vc): Descriptive narrative of potential user needs

(2) Product attributes (Vprod): Describe the attributes of the product according to the classification description

(3) Consistency constraint (CR): Defines the percentage of user attributes within the consent range

(4) filter condition (CF): Defines which product should be selected under which conditions

(5) Product constraints (Cprod): Defines the current valid product classification

(6) Collect query: different from the above. A collection query is a database query that connects a selection of criteria to a collection

Example and similarity degree

The case-based recommendation method uses similarity to retrieve items, and the similarity can describe the degree of matching between the attributes of a narrative item and the needs of some given user. This similarity calculation, according to the different scenarios, there are some more general formulas.

Interacting with constraint-based recommender systems

Session Interaction Process

(1) Users to specify their own initial preferences

(2) When sufficient information about the user's needs and preferences is collected, it is provided to the user with a set of matching products, and the user can choose to ask the system to explain why a product is recommended

(3) Users may change their own needs

Technology to help you interact

These techniques, which help to recommend systems and user interaction, help to improve the usability of the application and achieve higher user acceptance in terms of trust, comfort and results of the recommendations.

(1) Default setting: Recommend default, select Next question

(2) Handling uncomfortable requirements and empty result sets

(3) Proposed changes to unmet needs

(4) Sorting based on item/utility recommendation results: It is important to sort the user's utility according to the item, because the first effect, the user will pay more attention to and select items at the beginning of the list. Such sorting can significantly increase the trust of the recommended application and the user's willingness to buy.

In the knowledge-based conversational recommender system, the ordering of items is based on the multi-attribute utility theory. According to each item on the user's effectiveness to evaluate.

Summary of user and constraint-based referral interactions

Users can resort to decision results and changes when they cannot find a solution.

The default value can give a reasonable candidate scheme to help understand the need, and its negative effect is to misuse the default value to manipulate the user.

The utility-based sequencing helps to sort the information units on the results page, changes the candidate options provided by the decision result and the change ingredient, and sorts the recommended item interpretations.

These concepts form a toolset that helps to better implement constraint-based recommended applications.

Interacting with instance-based referral systems

Similar to constraint-based recommendations, the early-morning case-based recommendations are also purely query-based methods that users need to specify (often repeatedly specifying) their needs, knowing that the target item is found.

For non-professional knowledge, it is very difficult to understand the professional properties of items, based on which people have proposed a browse-based method to retrieve items. If the user does not know what they are looking for, this method navigates for it. Evaluation is a very effective navigation method, and is also a key concept based on the case recommendation system.

(1) Evaluation: Its basic idea is. The user specifies the requirements for the changes that are not met by the currently pending items (input items or recommended items)

(2) Mixed evaluation

(3) Dynamic evaluation

(4) Advanced Item recommendation method

(5) Evaluation of diversity

This recommendation process also has very many computational formulae. The calculation process is inside and can be studied in depth when used.

This recommended process. It is still very much like the commodity retrieval process.

Summary

Based on knowledge recommendation. The quality of the recommended application relies on the quality of the underlying knowledge base. The process is also very dependent on user feedback. Then, there are two types of:

(1) Constraint-based recommendations: When emphasizing constraint rules recommendation

(2) Recommendations based on examples: it is emphasized that, based on the user's original search results and stepwise evaluation (which can be understood as the choice of many other search criteria), we recommend a lot of other results.

Actually very much like conditional search.

"Recommender System"--based on knowledge recommendation