Bing’s blog introduced a new advanced search feature that allows programmers to locate code snippets. This article is notable for shedding light on Bing’s technology and hinting at its capabilities, which remains largely enigmatic to the SEO community.
1. Bing Utilizes Natural Language Processing
Bing employs natural language processing (NLP) to better understand user intent. NLP is a branch of artificial intelligence aimed at comprehending human language patterns.
What is Natural Language Processing?
Some in the industry doubt the application of NLP in search engines. This skepticism might be due to a lack of understanding, despite both Google and Bing being transparent about its use. A Microsoft document explains NLP as encompassing language knowledge (phonetics, semantics, morphology, etc.) and also involves text classification, machine reading comprehension, and question answering.
The Bing blog explains how they use NLP to interpret human inputs of code. The objective is to determine user intent and convert it into something the search engine can comprehend.
"To achieve this level of precision for query intent detection, Bing’s natural language processing pipelines for developers leverage patterns found in training data from developer queries collected over the years containing commonly used terms and text structure typical for coding queries."
In this context, NLP assists in interpreting human search queries by comparing them to patterns from previous queries. However, NLP encompasses more than this and Bing (like Google) likely utilizes its full spectrum (semantics, morphology, etc.).
2. Bing Uses Click Signals to Enhance Accuracy
"The system also leverages a multitude of click signals to improve the precision even further."
This likely refers to using click patterns to interpret user intent based on specific search queries. A consistent pattern can indicate a clear user intent, while inconsistent patterns suggest ambiguity, leading to varied search engine results pages (SERPs).
3. Bing Uses Upvotes for Ranking Forum Answers
This is a revealing insight into Bing’s ranking process. For coding-related queries, Bing pulls pages from forums and considers up-votes and down-votes as indicators of the accuracy of the answer.
"…the system then extracts the best matched code samples from popular, authoritative and well moderated sites like Stackoverflow, Github, W3Schools, MSDN, Tutorialpoints, etc. taking into account such aspects as fidelity of API and programming language match, counts of up/down-votes, completeness of the solution and more."
4. Does Bing Consider Moderation a Trust Signal?
Bing emphasizes effective moderation as a positive ranking signal:
"…from… well moderated sites…"
This suggests that moderation might be a factor in ranking user-generated content. A forum’s moderation level could be gauged by spam levels or the absence of obscenities or troll language.
5. Popularity and Authority of Answers
The blog post highlights popularity and authority as trust signals.
"…from popular, authoritative… sites…"
While this specifically refers to extracting answers from forums for code searches, it indicates what Bing looks for when ranking a site. Popularity and authority might be assessed through social media signals, links, and user engagement, though the specifics remain uncertain.
Takeaway: Understanding Bing is Essential
Historically, search marketers have found Google more approachable than Bing, partly due to the known importance of links in Google’s ranking system. However, this familiarity is arguably misleading as Google’s algorithms have evolved significantly.
Bing, perceived as mysterious, seems less reliant on links and more focused on user intent and content comprehension, based on Microsoft’s research topics. This makes Bing more resistant to link spamming, differentiating it from Google, and perhaps explaining why some sites ranking well on Google don’t perform as well on Bing.
As Google advances in understanding natural language and content semantics, it’s beneficial to understand Bing’s workings, as both search engines are converging towards similar goals.
Images by Shutterstock, Modified by Author
