Learning Search Engine Optimization: 5 Top Semantic search engines

Intermediary transaction SEO diagnosis Taobao guest Cloud host technology Hall

In search engine circles, "semantics" is a word with magical aura. The hype will make you wonder if this could be the culmination of a second search. The publicity made me feel a little skeptical, and I've been waiting for the technology to mature. This is the time, and I'm happy to introduce you to 5 top semantic search engines.

What is semantic search?

A semantic search engine attempts to interpret search results through context. It can automatically recognize the conceptual structure of text. For example, if you search for "elections", semantic search engines may get textual information that includes "voting", "campaign" and "ballot", but the word "election" may not appear at all in these sources. (Editor's note: That is, semantic search can interpret keywords and similar words, thus broadening the accuracy and relevance of search information.) )

In the semantic search process, there is an important component, that is, for the query information and the content of the network, the disambiguation (eliminate ambiguity). This means that search engines are processed by natural language--when you search for "Jaguar" (Jaguar Sedan), looking for a car or a big cat. Editor's note: The current popular search engine is basically the focus on the keyword, which keyword search more people, basically what kind of results will appear in the forefront. )

Here are some of the search engines you're going to introduce that use semantic analysis to filter and interpret data. However, you will also see that these search engines take different forms and show five different products.

When do you make semantic search engines?

Semantic search has the ability to improve the traditional web search experience, but it does not play a role of substitution. Most of the search problems on the network are very random browsing, the results of the requirements are not improving, so the usefulness of semantic search is not, but targeted research search, will benefit from the results of semantic search.

Now look at our list of the 5 most popular semantic searches to see if they will improve your search experience. Hakia is a general-purpose semantic search engine, unlike the powerset,cognition described below, whose search results are a corpus like Wikipedia's.

Hakia's search results are organized through tabs (tags): web results, reliable sites, images and news, and reliable sites that are certified by librarians and other information professionals invited by Hakia. (Editor's note: Librarians in the United States are very knowledgeable and educated information experts, each branch has specialized librarians, they are good at classification of information retrieval, every year in Washington, there are library association meetings, all over the country librarians will gather together to explore new information classification, technology search. Many of today's technologies and search patterns are the first to initiate and develop. )

Compared to some simple search questions (typically those that are popular and lack ambiguity), Hakia will give a similar resume. This is a collection of related information based on topics. Each "Resume" is indexed by a link to some information and gives a quick reference to the page for immediate presentation.

The content of these "resumes" will vary according to the nature of the search problem (e.g., personal files, bibliographies, big events, can be personal, government, economy, culture, or information about the country). This "Resume" for me is my favorite Hakia features, especially when I search some topics.

Hakia usually recommend some related searches, which is a great support for research.

For example, if I were to search for Obama, HaKia would recommend whether I was interested in his wife, Hillary, Democrat, Republican Sarah Palin, McCain, etc.

But there are some problems Hakia will also give the result of poor quality, but Hakia or in the test phase, I believe there will be rapid improvement. Ensebot is a search engine that summarizes the results of a refined digest form. It tries to understand the meaning of the interface of these search results. From this point of view, it uses text mining, trying to analyze Web pages and identify their key semantic concepts.

Sensebot helps you quickly grasp what the relevant information is in this way. From this point of view you don't have to look for a lot of web pages, or even comb the definition of an incomplete expert in the resulting results. )

This refining summary is useful for digesting your search topics, bringing together the relevant and important content of various search results. It also provides you with a label cloud of various related concepts and a form that is considered to be a definition summarizing your search questions. Each statement also includes a link to their source.

Not all summaries are widely or easily understood, but are highly likely to be improved. Hakia,sensebot are still in the testing phase. These exciting and advanced technologies are progressing with each passing day.

Powerset is now not a regular web search engine for Internet. It can play the largest xiao in a relatively small, results-capable corpus, and the technology provides comprehensive access to information about me. You can test on Wikipedia, but you'll find that Powerset is better at this search, and the way it constructs information and presentation for research purposes is based on the great improvement of Wikipedia. You can enter keywords, phrases, or a simple question in the search field. Powerset usually directly answer questions in the search results interface. One of my favorite features is powerset the information in multiple articles.

"Factz" is a box that often appears in search results, and it is a reference material that includes suggestions based on available information. For example, when I search Obama, Powerset provides links to information on some Obama's views on Robert Gates, the Middle East, Pakistan, and trade. Clicking on a link in the search results will show a small dialog box with a quote from Obama's statement in the article, as well as a link to the text of the quote.

DeepDyve (Editor's note: DeepDyve is one of the most promising semantic searches in the world, and the editor noticed the company last year and contacted several programmers of the company and found that a lot of Chinese work here.) )

DeepDyve is a very powerful, professional research tool that provides free service to the public. (Editor's note: DeepDyve's search service is free, but if you want to read the published academic journals, pay for them.) )

This is a research engine that helps you get into the "deep web" of professional content: information that is not indexed or included in the Internet by traditional search engines (such as some professional databases, periodicals, etc.). Scholar, researchers, students, technologists, business and other DeepDyve users and other information consumers can search for Wikipedia and "deep web" information in the following categories: Life sciences, medicine, physical sciences, humanities and social sciences, business and finance, patents, law, clean energy technology, electronic engineering.

Search engines on Web sites often rely on Boolean or hard-coded classifications (Editor's note: Logical computing language), which has a threshold and a certain learning curve, and it is difficult for people to use them. DeepDyve is currently a service search engine with a simple interface to present these valuable sources of information. Your search question can consist of any single word or even a 25000 word problem. The results of this search are presented in a complex way, including many advanced options: refining, separating, and storing your search. The search results are relatively easy to navigate, even with complex types.

Cognition's search business is based on semantic maps, and 24 years after it was founded, Cognition was able to claim to be the most complete and most refined English search map available today.

You can use cognition technology to search for information in the four carriers below.

· Public.Resource.org (Currently there are 675,704 federal case law files, 1,858 volumes, all of which are XHTML-formatted files.) The information released has been made up of the decision of the High Court and the Court of Appeal since 1950.

· MEDLINE (medical literature analysis and online retrieval System) Summary library: This database abstracts from life sciences and biomedical science information in the world Literature database. It includes information on medicine, nursing, medicine, dentistry, veterinary medicine, and health care, as well as information on areas not directly related to medicine, such as molecular evolution. (There are currently 18,005,903 files).

· All English versions of Wikipedia content

· The most comprehensive new English translation content, including text and translators ' notes such as the Gospels thegospels of Matthew, Luke, John and Mark.

We tested cognition in Wikipedia. In this large text database, cognition is particularly good at classifying the meaning of complex issues.

· Words: For example, "the old chapel and the old temple" undefined

· Meaning distinction: "Striking workers" contrasts "California's oil gold" undefined

· Category: For example, "Indian tribes in Latin America" or "disease of North American trees" undefined

The technology used by these cognition to solve search problems is impressive, cognition gives users a friendly way to control meaning and classification more meaningfully. Undefined, I think my statement of cognition search results is not perfect, but I earnestly hope that conition team can also learn from Hakia or powerset experience. Undefined