*~~Search engine question~~*

[labrynth_of_fire]

Would it be possible to make a program that grabs the key words a search engine is looking for, and then feeds it back to the engine? (like if I typed monkey into the search bar and clicked search, the program would tell the engine that my website had the key word "monkey" ).

[minstrel]

I'm not sure I understand the question: How would you know what someone was feeding to a search engine unless that search engine had already brought the person to your site?

Maybe I'm missing something here...

[anuj_pandit1]

Hello labrynth_of_fire,
I hope you are doing well,

When any body will search of monkey into the search bar and clicked search, then your program should track the Search Engine Spider or Robots that Seacrh engine is using then to take out the result of the required query, then your program should start that what the keyword is using by the user.
Track this Keyword and your program will tell to the Sspider or Robots that your website is perfect with rules and regulation and it should get the prority to index in top 10 or top 25.

So this way you can creat a simple program to fetch the online query in search engine.

I hope labrynth_of_fire it will help you...

Regards
Alok Kumar Upadhyay

[elearning]

Along the same lines - how do companies like WordTracker build their databases? I am "guessing" that they've built relationships with the other meta-engines out there to capture their keyword searches.

Would this be a correct assumption?

Thanks,

Mohammed

[anuj_pandit1]

Mohammed Wrote

Along the same lines - how do companies like WordTracker build their databases? I am "guessing" that they've built relationships with the other meta-engines out there to capture their keyword searches.

Would this be a correct assumption?

-------------------------------------------------

I Agree with UUU Mohammed because you have need to develop program like Word Tracket that will identify that the maximum hit to which keyword and when it will get the Keyword of your website then it will tell to the search engine that this website is perfect to index.

I hope it will help you........

Regards
Alok Kumar Upadhyay

[elearning]

Thanks - I also found somewhere else that they use MetaCrawler.com and Dogpile.com results to build their database.

I've used their free service - and while it gives some good information, if you were to brainstorm and look at keywords that describe your business I think you'll get what Wordtracker gives you. Using an on-line thesaurus also helps when looking for keywords.

Mohammed


Mohammed Wrote

Along the same lines - how do companies like WordTracker build their databases? I am "guessing" that they've built relationships with the other meta-engines out there to capture their keyword searches.

Would this be a correct assumption?

-------------------------------------------------

I Agree with UUU Mohammed because you have need to develop program like Word Tracket that will identify that the maximum hit to which keyword and when it will get the Keyword of your website then it will tell to the search engine that this website is perfect to index.

I hope it will help you........

Regards
Alok Kumar Upadhyay

[elearning]

I should also add to that I found a cool website the other day called hitbrain.com - gives you similar information as WordTracker - but is FREE! I thought the info. was great.

Check them out...

Mohammed


Thanks - I also found somewhere else that they use MetaCrawler.com and Dogpile.com results to build their database.

I've used their free service - and while it gives some good information, if you were to brainstorm and look at keywords that describe your business I think you'll get what Wordtracker gives you. Using an on-line thesaurus also helps when looking for keywords.

Mohammed


Mohammed Wrote

Along the same lines - how do companies like WordTracker build their databases? I am "guessing" that they've built relationships with the other meta-engines out there to capture their keyword searches.

Would this be a correct assumption?

-------------------------------------------------

I Agree with UUU Mohammed because you have need to develop program like Word Tracket that will identify that the maximum hit to which keyword and when it will get the Keyword of your website then it will tell to the search engine that this website is perfect to index.

I hope it will help you........

Regards
Alok Kumar Upadhyay

[labrynth_of_fire]

thats not exactly what I meant, what I mean is can you make a program that grabs the word the web crawler is looking for, and tells it that its on the page you want to show.

[elearning]

Sorry, got off topic there.

Yes, you should be able to - though I'm not that skilled at web development. Web stats packages like AWStats and others do this.

I found this additional information on another forum:

You can use $_SERVER["HTTP_referer"] to get the referrer, you'll need some regex to extract the search query.

Then someone else said this:

It's going to take more work that you guys make it sound, but here's essentially what you need to do :
Try this on for size :

if(!isset($HTTP_referer)){
print "General message goes here.";
}
else{
$query = split("[?]",$HTTP_referer);
$div = split("[&]",$query[1]);
foreach($div as $var){
if(preg_match("/q=/",$var)){
$var = ereg_replace("q=","",$var);
$var = ereg_replace("+"," ",$var);
print "So, you're looking for info about $var";
}
else{
}
}
}


And finally:

Maybe you could also change:

if(!isset($HTTP_referer)){
print "General message goes here.";
}

to something like:

if(!isset($HTTP_referer))
{
print "<script TYPE=\"text/javascript\">\n";
print "if(document.referrer)\n{\n";
print " var args = new Object();\n";
print " var query = document.referrer.substring(1);\n";
print " var pairs = query.split(\",\");\n";
print " document.write(\"So, you\'re looking for info about \");\n";
print " for(var i = 0; i < pairs.length; i++)\n {\n";
print " var pos = pairs[i].indexOf(\"=\");\n";
print " if(pos == -1){continue;}\n";
print " var value = pairs[i].substring(pos+1);\n";
print " document.write(unescape(value));\n }\n}\n</script>\n";
}

Whoops - one more:

$referer = strtolower($_SERVER["HTTP_referer"]);
if (strstr($referer,"dog"))
{
echo("<h1>Dogs simply love our dog food!</h1>");
}
else if (strstr($referer,"bee"))
{
echo("<h1>Bees make a beeline for our bee food!</h1>");
}
else if (strstr($referer,"cat"))
{
echo("<h1>Cats can't get enough of our cat food!</h1>");
}
else if (strstr($referer,"hamster"))
{
echo("<h1>Our hamster food makes your hammy happy!</h1>");
}
else
{
echo("<h1>If your pet could order pet food online, this is where they'd come!</h1>");
}

Heck - here is the url.

http://www.webmasterworld.com/forum88/1569.htm

Enjoy!

Mohammed

thats not exactly what I meant, what I mean is can you make a program that grabs the word the web crawler is looking for, and tells it that its on the page you want to show.

[minstrel]

Yes, you should be able to - though I'm not that skilled at web development. Web stats packages like AWStats and others do this.

I found this additional information on another forum:

You can use $_SERVER["HTTP_referer"] to get the referrer, you'll need some regex to extract the search query.

Then someone else said this:

if(!isset($HTTP_referer)){
print "General message goes here.";
}
else{
$query = split("[?]",$HTTP_referer);

(snip)

But again - unless I'm missing something here, by the time you can extract this information, the search query has already found your page, right?

I think the original poster is asking how to determine what people are searching for on the internet so he can then tell them to look at his page for those keywords - it sounds like spoofing to me but I'm still a bit confused about this whole thread.

[elearning]

Heh - well that's what I thought and that's why I suggested WordTracker or HitBrain.

I don't know of any way to "grab" what keywords are being used on search engines. Somehow I don't think Google will give you access to their DB for free - though you could get a AdWords account and search on phrases being used to better target your keywords.

I also found https://adwords.google.com/select/tools.html.

Mohammed


Yes, you should be able to - though I'm not that skilled at web development. Web stats packages like AWStats and others do this.

I found this additional information on another forum:

You can use $_SERVER["HTTP_referer"] to get the referrer, you'll need some regex to extract the search query.

Then someone else said this:

if(!isset($HTTP_referer)){
print "General message goes here.";
}
else{
$query = split("[?]",$HTTP_referer);

(snip)

But again - unless I'm missing something here, by the time you can extract this information, the search query has already found your page, right?

I think the original poster is asking how to determine what people are searching for on the internet so he can then tell them to look at his page for those keywords - it sounds like spoofing to me but I'm still a bit confused about this whole thread.

So like ^ isn't that just keyword hunting??

My brain hurts. Enough diversions. I'm going back to work. :-)

Mohammed

[mikmik]

minstrel's point is 100% correct here:
elearning wrote:
Yes, you should be able to - though I'm not that skilled at web development. Web stats packages like AWStats and others do this.

I found this additional information on another forum:

You can use $_SERVER["HTTP_referer"] to get the referrer, you'll need some regex to extract the search query.

Then someone else said this:

if(!isset($HTTP_referer)){
print "General message goes here.";
}
else{
$query = split("[?]",$HTTP_referer);

(snip)


But again - unless I'm missing something here, by the time you can extract this information, the search query has already found your page, right?
I think a typical scenario is - type 'monkey' in google, press 'enter' or 'google search' and then results are displayed like this "Results 1 - 10 of about 5,060,000. Search took 0.14 seconds",
(that is what I got for monkey) - the point minstrel makes being that your "highly illegel" keyword spoofing program has 14 one-thousandth's of a second
to read the 'searcher's' input( maybe a "keylogger" trojan or some advanced 'packet sniffer/analyser' ubiquitously monitoring all internet traffic[best possible case scenario]) and then infiltrate google's own 'database query' and insert a plausible page result/description to be displayed like so -
Webmonkey: The Web Developer's Resource
... 6 Oct 2003. Today's Monkey Bite ... If you see strange bursts of color
in the sky tonight, don't worry. Nobody spiked your triple-tall! ...
Description: A how-to guide for web developers. Includes resources for design, multimedia, e-commerce, programming.
Category: Computers > Programming > Internet > Resources
hotwired.lycos.com/webmonkey/ - 23k - Cached - Similar pages ...

So the result would be that even if you were the most brilliant 'evil genius' cracker using the fastest, $200,000 supercluster supercomputer (able to leap 7 layer's of the OSI model in a singal bound), when the 'surfer/searcher' went to your page and found the only "monkey" around was the 'wrench' thrown into their efforts at getting information relevent to their initial query,and instead got their eardrums and sound system annihilated by the most egregious and LOUD website 'design', and I use the term 'design' in it's most liberal and forgiving interpretation, since 'FrontPage' was conceived thus revealing the hiding place of the curiously short-sighted, but never the less, brilliiant but misguided evil genious himself, they would most likely notify the 'authorities' as to your endeavers and location, thereby setting the record for the shortest turn-around time between the application of a computer based attack and the collecting of a $250,000 reward for information leading to the arrest blah blah etc of a criminal!

I guess the question is "WHY?!".

[mikmik]

In any event, I came upon a handy little app from 'AnalogX', therefore it is well written freeware, called "KEYWORD LIVE" that may offer insight into the idea of 'real time keyword usage'.
Find it here, http://www.analogx.com/contents/download/network/keyword.htm.

It is really quite intersting, after running it for long enough [1/2 an hour} to get 7,000 search queries, I had 'rectal enlargement' listed at #12 with about 15 searches. I think 'MP3' was the most at about 23 or something.

Also 'Keyword Extractor' is another utility of note available there.

[lmims]

Would it be possible to make a program that grabs the key words a search engine is looking for, and then feeds it back to the engine? (like if I typed monkey into the search bar and clicked search, the program would tell the engine that my website had the key word "monkey" ).

What you could do is build a list using the keyword suggestions then run those words through a thesaurus and then have the thesaurus word go through the suggestion process ad nauseum.

Or you could use Markus Allen's Keyword sleuth that automatically acomplishes this. Here's a link:http://www.armchairmarketer.com/sleuth/

Please note that Mark is changing servers and the DNS has not fully propigated. What that means is it may be tomorrow before you can give it a try.

[spiderbait]

Is it possible that the question behind this thread is rooted in a misunderstanding of the behaviour of search engines?

My first thought on reading this was that labrynth_of_fire mistakenly believes search engines scour the web in real-time in response to a search query.

In case that's the source of the confusion here, I'll just mention this briefly:

Of course they don't and they deliver their search results from huge databases that have taken months and years to develop. But if they did, then labrynth_of_fire would have a chance of building a program that would wait until a spider came by looking for a particular search term and then feeding that term back to the engine.

[TrafficProducer]

Warning.

Sounds simular to Cloaking This will get you banded.

More info:-
http://www.solutions.ukdots.com/advertising_methods.html

Using Dynamic Web pages could produce pages depending on Keywords on your own site, this is a normal search.

[carbonize]

You could redirect search engines by looking at the useragent string and if it's a known bot feed them a keyword loaded page rather than the real page. This way you wouldn't have to worry about your page looking ugly and not making much sense.

My PHP is non-existant but it would be something like


If (useragent='googlebot') then include('robots.html') else include('normal.html');

[RISTMO]

Spiderbait hit it right on (Minstrel was also on). I understand what you want, Labryinth, and it's not possible. When someone searches in Google, they aren't searching urls of every website that Google has the url of. Instead, they're searching a cached version of the page at each url. The problem is that since they're searching a cached version which is stored in Google's databases, you have no way to get a program to run in there, since it's going directly from Google's search page to its database and back. If it were crawling your site, it would be simple, but the problem with that is that when Google crawls your site, it doesn't crawl it with any specific keyword in mind. So it's really not possible at all.

Rick

[labrynth_of_fire]

Warning.

Sounds simular to Cloaking This will get you banded.

More info:-
http://www.solutions.ukdots.com/advertising_methods.html

Using Dynamic Web pages could produce pages depending on Keywords on your own site, this is a normal search.
Is that what I am thinking of (Cloaking?)

[carbonize]

No my last post was cloaking, your idea would involve hacking into the google servers and making it serve your page up for every request it recieves which would not only get you banned but probably also a visit from Officer Dibble or maybe Mr. Smith.

[labrynth_of_fire]

No my last post was cloaking, your idea would involve hacking into the google servers and making it serve your page up for every request it recieves which would not only get you banned but probably also a visit from Officer Dibble or maybe Mr. Smith.
Oh Ok I don't know how to hack and don't want to either....But that idea is similar to what i was thinking of.

[DrTandem1]

What's the point? Let's say you had such a program, and any keyword that someone entered would send them to your site. That's your goal, isn't it? Besides the fact that 99.99999999999999% of the searchers would not find on your site the content that they actually were trying to find, your server(s) would crash.

Anyway, this is not how a search engine operates. They crawl the internet and index sites(create/modify THEIR data base) using a variety of elements, not just keywords. In other words, the search engine is not looking at your site actively (real-time) for the searcher. It is researching its existing data base. While creating this data base, the search engine is not searching for keywords. It's not really searching for anything during its crawls.

If it were operating in this manner, a dictionary site would always be the number one result of any search.