Query classification; understanding user intent
http://vervedevelopments.com/Blog/query-classification-understanding-user-intent.html
What exactly is the function of a search engine? In simplest terms it acquires, stores and returns information (from the web). Ok, simple enough. But we're talking about people here, people seeking information. How they interact with the search engine is often a huge problem. What is the user intent? A good way of starting to pick apart that puzzle is by classification of query types. And that's what we're going to be looking at today.
To understand what a user truly wants when searching for something you'd need to ask each user what it is they are after. While that works for the offline mom and pop store, it isn't at all feasible for a search engine. Thus an automated approach needs to be taken. What's more limiting, is that they have to infer intent from very few words and little in the way of (explicit) interaction.
Why would that be important to SEOs? Well, that's even easier. If we understand how people search for things, we are in a far better position to actually target our programs to ensure the highest, most relevant, levels of traffic for our sites and our clients. Understanding how search engines attack the problem, can be VERY useful in our own targeting and programming.
The basics of query classification and beyond
One of the reasons a search engine looks at classifying queries is to better understand user intent. To do that they will look at the search task process as such;
- Enter query
- Retrieve results
- Scan results
- View results (actual pages returned)
- Refine query (if needed)
Interestingly searchers have certainly evolved over the years beyond mere information needs, into commercial and navigational (seeking known entity) as well. We also can consider that, unlike say.. a library, web searchers are looking for a wide variety of mediums (text, images, multimedia) as well as from various locales (work, home, mobile). Search engines perform social networking functions. They Act as dictionaries, spell checkers and thesauruses.
This is why classification has become more and more important to search engineers over the years. Understanding, as close as possible, the intent, is paramount. The interesting part is the ever changing landscape of exactly what the intent is.
Some studies make the case that users are prone to a higher level keyword approach to simply get near the vicinity, preferring to click through at that point and search the local site for the exact information need. These have been referred to as 'teleporting queries'.
Classification of query types, among SEOs at least, have generally come in three flavours;
- Informational
- Transactional
- Navigational
But, these are simply broad categorizations that we should play hard and fast with. Many queries fall into more than one category and that's actually quite important for SEOs to understand. The reason we care about these is that they play a strong roll in keyword research and ultimately targeting and content programs.
The following table gives you a sense of the various classification types in this area (click for full size);
A different mind set
So let's go beyond the traditional understanding of query types. We have looked at the core types so far, but another paper I came across broke things down a little differently. The reason I decided to bring this up is because we need to understand things aren't always the same.
Here's a chart from the paper which helps understand this approach;
1. Navigational |
My goal is to go to specific known website that I already have in mind. The only reason I'm searching is that it's more convenient than typing the URL, or perhaps I don't know the URL. |
aloha airlines duke university hospital kelly blue book |
2. Informational |
My goal is to learn something by reading or viewing web pages |
|
**2.1.1 Directed |
I want to learn something in particular about my topic |
what is a supercharger 2004 election dates |
2.1.2 Open |
Closed I want to get an answer to a question that has a single, unambiguous answer. |
baseball death and injury why are metals shiny |
2.2 |
Undirected I want to learn anything/everything about my topic. A query for topic X might be interpreted as "tell me about X." |
color blindness jfk jr |
2.3 Advice |
I want to get advice, ideas, suggestions, or instructions. |
|
2.4 Locate |
My goal is to find out whether/where some real world service or product can be obtained |
pella windows phone card |
2.5 List |
My goal is to get a list of plausible suggested web sites (I.e. the search result list itself), each of which might be candidates for helping me achieve some underlying, unspecified goal |
travel amsterdam universities florida newspapers |
3. Resource |
My goal is to obtain a resource (not information) available on web pages |
|
3.1 Download |
My goal is to download a resource that must be on my computer or other device to be useful |
kazaa lite mame roms |
3.2 Entertainment |
My goal is to be entertained simply by viewing items available on the result page |
xxx porno movie free live camera in l.a. |
3.3 Interact |
My goal is to interact with a resource using another program/service available on the web site I find |
weather measure converter |
3.4 Obtain |
My goal is to obtain a resource that does not require a computer to use. I may print it out, but I can also just look at it on the screen. I'm not obtaining it to learn someinformation, but because I want to use the resource itself. |
free jack o lantern patterns ellis island lesson plans house document no. 587 |
This approach certainly helps to give you the idea of what search engineers are looking at. Does it really matter if it is the classical informational/transactional/navigational or resource? Of course not. What we're doing here today is seeking to get a feel for how classification may be done.
To a search engineer, on the larger level, there are two simple aspects to a query; intent and satisfaction. Each person using a search engine has a goal and classification helps break down these goals into bite sized pieces.
Not to be taken too seriously is some of the data this particular group found in their research;
I say not to take it to heart because as we all know a single data set never tells us the entire picture. This was actually taken from Alta Vista data, soooooo... take it for what it is.
Associating Goals with Queries
If we consider goals as understanding intent, then we can break associations into two areas familiar to most of the search geeks reading my ramblings over the years; implicit and explicit.
Explicit – in most cases these would be navigational or more simplified queries.
Implicit – would be less obvious queries or system elements such as Google's [I'm feeling lucky] feature.
Another common element we see in query classification is building machine learning approaches based on training sets. As you'd imagine, the larger the data set of query data you have, the better associations you can make between queries and intent/satisfaction. Just because behavioural data may not be overly-valuable for ranking elements, doesn't mean it's off the table altogether.
In the paper I cited earlier, they talk about using behavioural data to seek out telling signals the user might give;
- the query itself
- results returned by the search engine
- results clicked by the user
- further searches or other actions by the user
There are actually other actions that can be tracked such as;
- dwell time (on page clicked on)
- scrolling
- page depth
- saving the page to favourites
- Explicit SERP actions (such as Google's +1)
You get the idea. Behavioural data, on a large scale, can bring a great deal of data to further understand the goals and potential intentions of users through implicit and explicit data. We can also see this in recommendation engine elements (Google Suggest, refinements etc.).
What Can SEOs Learn From Query Classification
To begin with, let us look at the core goal of classification; assessing user intent. That of course should be obvious as far as why we, as SEOs, would want to also understand this. There are no tools out there that really give us this. Which means, to some extent, we have to look at potential query spaces and establish what user goals we're trying to service. If you use the classic informational/transactional/navigational approach or the above 'resource' model, is inconsequential. What we need to do is align targeting and content programs to best serve these needs.
When do we look at it? For the most part understanding query classification plays into one of the first elements of an SEO program; keyword research. Out SEO programs live and die from the efficacy of the keyword research. Keyword research is focused on matching user intent with our (ranking) targets.
It should be noted that most queries are informational in nature. Even quasi-classifications such as seeking information on a product prior to purchase. In fact, much of the research seems to show that navigational queries are often seeking information about a product and the query is often refined to reflect this. As such we must consider having a content program that reflects this.
Below are some tables from a recent research paper that shows the percentages of each (in the traditional model) for various topic areas.
The main goal here today was to give you a sense of how search engines are dealing with this so that you can start to adapt your own keyword research and content programs accordingly. If you're interested in more detailed planning, be sure to sign up for the SEO Training Dojo as we will be putting this (and much more) into a keyword research section being posted in the next week or so.
I hope you enjoyed the ride... I know I did.
Papers used/cited in this post;
Understanding User Goals in Web Search
Classifying Web Queries by Topic and user intent(direct link to PDF)
Clustering Query Refinements by User Intent
Query classification; understanding user intent的更多相关文章
- 计算广告(5)----query意图识别
目录: 一.简介: 1.用户意图识别概念 2.用户意图识别难点 3.用户意图识别分类 4.意图识别方法: (1)基于规则 (2)基于穷举 (3)基于分类模型 二.意图识别具体做法: 1.数据集 2.数 ...
- 基础总结篇之九:Intent应用详解
看似尋常最奇崛,成如容易卻艱辛.北宋.王安石 看似普通的事情其实最不同寻常,并不是简简单单就可以做好的:成功看起来似乎很容易,而成功的过程却充满着艰辛. 对于我们认为很普通的事情,不屑一顾,就永远不会 ...
- android intent打开各种文件的方法
android intent打开各种文件的方法 1./** * 检测是否安装了某个软件 * * @param pkgName "com.bill99.kuaishua" ...
- SharePoint 2013: Search Architecture in SPC202
http://social.technet.microsoft.com/wiki/contents/articles/15989.sharepoint-2013-search-architecture ...
- android 打开各种文件(setDataAndType)转:
android 打开各种文件(setDataAndType) 博客分类: android-->非界面 android 打开各种文件 setDataAndType action动作 转自:htt ...
- Android(java)学习笔记143:android提供打开各种文件的API接口:setDataAndType
android 打开各种文件(setDataAndType) private void openFile(File file){ Intent intent = new Intent(); inten ...
- 微软职位内部推荐-Senior Software Engineer-DUT
微软近期Open的职位: Document Understanding and Task (DUT) team in STCA focuses on semantic understanding an ...
- Android 4 学习(18):搜索
参考<Professional Android 4 Development> 搜索 通过下面这几种方式可以给应用程序添加搜索功能: Search Bar Search View Quick ...
- Android(java)学习笔记86:Android提供打开各种文件的API接口:setDataAndType
1. Android 打开各种文件(setDataAndType) private void openFile(File file){ Intent intent = new Intent(); in ...
随机推荐
- 简单的表单验证插件(Jquery)
在做web开发的时候经常遇到表单验证问题,表单验证一般有客户端验证和服务器端验证,这个验证插件仅仅能满足我的项目中的基本需求的. Validate_Tools.js function Validate ...
- 增强for循环用法___ArrayList数组实现使用下标最好,LinkedList使用增强型的(转载)
总结: 1.For-Each循环的缺点:丢掉了索引信息. 当遍历集合或数组时,如果需要访问集合或数组的下标,那么最好使用旧式的方式来实现循环或遍历,而不要使用增强的for循环,因为它丢失了下标信息. ...
- jquery实现多行文字图片滚动效果
今儿分享一个jquery实现多行滚动效果. 我看一些论坛网站上面,公告处用的较多. 代码如下 复制代码 // 多行滚动(function($){$.fn.extend({Scroll:function ...
- 关于Java中计算日期差值不准确问题
1.字符串日期相减 如:2016-4-1,必须先将此字符串转成Date对象,并且, 格式必须为:yyyy—MM—dd HH:mm:ss. 如果不转就直接计算(2016-4-1)两个这样的日期,则误差 ...
- c#中sqlhelper类的编写(二)
上一篇文章讲了简易版的SqlHelper类的编写,我们在这里就上一篇文章末尾提出的问题写出解决方案. sql语句注入攻击已经是众所周知的了.我们如何在C#中保护自己的数据库不被这样的方式攻击呢? 不用 ...
- java利用反射绕过私有检查机制实行对private、protected成员变量或方法的访问
在java中,如果类里面的变量是声明了private的,那么只能在被类中访问,外界不能调用,如果是protected类型的,只能在子类或本包中调用,俗话说没有不透风的墙.但是可以利用java中的反射从 ...
- java中对集合对象list的几种循环访问
java中对集合对象list的几种循环访问的总结如下 1 经典的for循环 public static void main(String[] args) { List<String> li ...
- 修改ubuntu按电源键触发效果
GUI内终端执行如下指令,或者加入开机启动脚本内 gsettings set org.gnome.settings-daemon.plugins.power button-power shutdown ...
- float闭合(清除浮动)和CSS HACK
一.float 闭合(清除浮动) 将以下代码加入Global CSS 中,给需要闭合的div加上 class="clearfix" 即可,屡试不爽. <style>.c ...
- Linux5.8下安装PhpMyadmin无法关联php-mcrypt问题
一.yum安装php-mcrypt ##发现没办法安装 原来CentOS 官方默认不在对mcrypt模块 进行支持,所以必须另想办法折腾了2个小时总算搞定,这里主要使用了Fedora的扩展库, E ...