Editorial archive of Creangel. The references, figures and conditions correspond to the original publication.
As the data-based era gives way to an information-based economy, the relevance of data is essential for obtaining practical information, as well as relevant comments, especially from advanced users, who will also play an important role in making information impact. This article explains the concept of an appropriate feedback model and why organizations should be interested in this relevant context.
What is the relevance of the search? Suppose they ask someone to provide an information system on a topic. This topic can have many facets and can obtain information from different sources. If you deal with this person or system on a regular basis, you may want to tell them that there are only certain aspects and therefore certain types of information that apply to you and hopefully you will only get the most relevant answers from them in the future. Then the relevance of the search is the ability of the system to get its relevant comments to adjust the results of your future query so that they are the most sought after and important for each user. The system performs and automates this task by adjusting the weight of certain terms and their equivalents (i.e. terms with the same or similar meaning) in the data it processes.
An example for us to be clear, let's think that within the company there are several users with various queries and various topics. One of the users asks a question about a specific topic, the next time you ask, you will get information related only to the latter and possibly related topics such as your search.
For a person and a consultation, it seems quite simple. But now imagine that you have tens of thousands of colleagues and thousands of topics to cover. This is when the relevance of the search uses machine learning algorithms to discover not only the preferences of each individual, but also groups of people with similar interests, similarity of documents, etc. It provides continuous and automatic feedback from users on the relevance for other documents, queries and people.
The main advantage of the relevance of searches is that it allows users, especially advanced users, to influence the appropriate dependencies for their environment without requiring IT to implement dependency rules based on specific user groups. This allows administrators to configure specific organizational users and precise compliance enhancement factors through the configuration.
Relevance can also significantly improve man-machine interaction. As the relevance of certain content increases significantly with relevance feedback, the user experience begins to feel much more “conversational”, that is, offer from one to three suggestions as “answers” to a query than a traditional search engine with an interface that returns a list of documents in response to a query. It also provides a way to find information about each person's experience that best answers the question. Take the example of a customer service agent looking for an answer to a customer's question about a product using a product name or code. In this case, the agent will obtain several documents, including spare parts catalogues, practical information, product specifications, packaging information, marketing materials, etc. All this information is relevant, but only certain information can help the agent answer the question.Customer problem. Thanks to the relevance, the agent will immediately see information that he has already seen in previous searches for similar things, because the relevance takes into account the “click action” of the user and applies a small increase of relevance accordingly.
Relevance can also rearrange results by observing (over time) information that other agents have taken the time to discover, even as they delve into the list of results to find relevant information. Organizations that try to make the most of the relevance, will configure it so that the interactions of experts with the system of more impact to the important content and even avoid that inaccurate information appears in the list of results.
As seen in the above example, relevance in search provides a collaborative method for reordering search results. It is not a method of labelling or sorting, both can be done at the time of index (extraction of source metadata, extraction of entities by natural language processing) or later (classification by machine learning algorithms such as grouping, calculation of similarity, etc.). Relevance could be said to be a smarter approach that directly incorporates human decision-making and provides the information that best suits user questions.
In conclusion, as information-driven organizations strive to achieve greater precision for end users seeking knowledge, the ability to automatically take advantage of relevant feedback from users, especially advanced users, is increasingly important for optimal business performance and the relevance of searches plays a key role in achieving this goal.
Author: Creangel Ltda