Frustrated Domain Searchers
If you have tried to register a domain name lately, you’ve probably been quite frustrated with the whole process. The reasons are simple – good domain names are scarce and the popular ones have long been claimed.
Traditionally, when you needed to register a domain name, you went to a credited domain name registrar’s website, such as GoDaddy. Say you want to register the name, “filmratings.com.” You would type “filmratings.com” in GoDaddy’s search box and click the search button. Whoops, it’s already taken! Then you think, what about movieratings.com? Nope. Let me try movierankings.com. Nope. How about movierating.com, filmrating.com, onlinefilmratings.com? All taken. You end up sitting hours in front of you computer before finding a domain name you really like – if you’re lucky.
Domain Suggestion not Easy for Computers
So you wonder, wouldn’t it be nice if there were a tool that can automate the process and make this whole ordeal less painful? If this strikes a chord, you are not alone. Every month, 35,000 searches are made in Google for “domain suggestions” and “domain name suggestions.” This is a lot of frustrating searches.
Computers can help in the process of finding a domain name, but it is not always easy. Humans have the ability to reason and manipulate language. While searching for an alternative to “filmratings.com,” for example, a person can substitute each word with their synonyms, “movie” for “film,” or “ratings” for “rankings.” Not an easy thing for computers.
Humans are also able to grasp a word’s different shades of meaning. For example, “film” can mean a movie (as in a film festival) or the medium photographers use to shoot photos (as in camera films). The meaning changes with context, but a computer is not able to understand context; thus, they ignore the slight variations in subtext.
Finally, humans are able to make word associations that computers cannot. For example, “credit card debt,” “low rate credit card,” “get a credit card” are good expansions of “credit card,” while “sexy credit card,” “strong credit card” are not. Computers cannot replace a person’s linguistic and world knowledge-but it doesn’t stop the geeks from trying.
A Brief History of Domain Search Tools
In the beginning, there were web tools – usually called name generators or name spinners – that could use language rules or word combining techniques to generate a list of names. These rules, for example, changing singular nouns to plural ones or adding common prefixes or suffixes such as “my,” “online,” or “best,” are usually very simple and easy to craft. The disadvantage of using these rules, however, is that the number of suggestions offered by the computer is very limited. This approach, although naive, has been adopted by most domain name registrars because of its simplicity.
Then, in a means of improving the methodology, more sophisticated language technologies were adopted. By using technologies such as synonym substitution and statistical word collocations, these systems, best represented by DomainsBot, can generate hundreds of suggestions for any given keyword, greatly expanding the scope of the technology. The problem to this approach, however, is that many suggestions don’t make much sense and thus have little value. For example, the top five suggestions for the keywords “domain suggestions” from DomainsBot are:
- Integral Domain Suggestions
- Frequency Domain Suggestions
- Domain Suggestions Objections
- Binding Domain Suggestions
- Exclusive Domain Suggestions
Clearly, this is far from perfect.
Enter Natural Language Processing (NLP). This technology has become the core technology and building block for many Internet companies, such as Google and Amazon. Sophisticated NLP technology has pushed the domain name search to a new high-at least, to a usable level. Within the last couple of years, two companies have used this system and made it work: DomainWhiz.net and DomainTools.
Here are the top five suggestions for the keywords “domain suggestions” from DomainWhiz:
- DomainNamesSuggestions
- SuggestDomainNames
- DomainNamesSuggestion
- DomainSearchSuggestions
- DomainSuggestionsTool
…and from DomainTools:
- URLSuggestions
- WebDomainIdeas
- DomainSparks
- SuperSiteIdeas
- DomainAlternativess
These suggestions (still available today, by the way) all make sense.
The two companies each have their weaknesses and strengths. In general, DomainTools’ suggestions are shorter, a plus. On the other hand, DomainWhiz’s suggestions tend to use keywords that would come naturally to a user. These keywords, with high SEO values, boost a website’s ranking in search engines’ returns, which leads to more visitors and more web traffic to the site.
How do the domain suggestion websites make money? DomainWhiz earns a commission by directing their users to GoDaddy to finalize the purchase. DomainTools’ revenue is obtained from user memberships. DomainTools shows the top 20 suggestions to the end user. If the user wants to see more suggestions than this, he is required to pay a $15 monthly membership fee. Although unconfirmed, DomainTools also appears to be a GoDaddy reseller.
Jerry Vasu, the author of this article, is a researcher specializing in natural language technology. His research interests include sentiment analysis, computer-aided language learning, and information retrieval.
Author: Jerry Vasu
Article Source: EzineArticles.com
Humorous photo captions
camera films, card, com, credit, credit card debt, domain, domain name registrar, domain name suggestions, Domain Names, DomainTools, example, google, name, ordeal, search, search button, subtext, Suggestions, word, word associations
