Thursday, March 24, 2011

Artificial Intelligence

















HAL 9000: It can only be attributed to human error. 2001 A Space Odyssey.





Lots of methods claim to include artificial intelligence in their systems, Robots, Chat Bots, Avatars and lots of assorted computer software programs. BUT none of them are – really intelligent.

Intelligence – cognisant of self and understanding of things in relationship to, information, analysis, knowledge and armed with this taking actions – decisions based upon knowing and estimations and predictions of ‘partial knowing’.





Dictionary definitions:
The capacity for learning, reasoning, understanding, and similar forms of mental activity; aptitude in grasping truths, relationships, facts, meanings, etc. the faculty of understanding.
knowledge of an event, circumstances, etc., received or imparted;

Also the potential to make decisions based upon partial knowledge to result in an outcome that did not previously exist.

These are signs of intelligence and as yet no computer system has exhibited such things. Also the dictionary omits to describe: cognisant of self

Search engines do not have such capacity – they have no capacity to be self aware of their functions: Google uses an algorithm and gives out multitudes of spurious (Spammy) results – Blekko is an attempt to reference information into much more useful categories with relationships.

And how do they do this:







             They do it with people building these SLASH-TAGS

The result makes the searches better, more accurate and much closer to the searchers need and with additional related data linkages, this cannot be achieved by Google, and BLEKKO is doing it with the assistance of people – a Crowd Source building model; people using their brains to intervene: adding the ‘KNOWLEDGE’.

So the way to make search engines and their future derivatives more intelligent is to use human intervention.

This is the very opposite strategy in the mainline areas of th A.I. development industry; who believe that top down computational algorithms will produce artificial intelligent systems: I say NO.

How can a stupid none aware mathematical engine generate the definition of intelligence as defined above: NO WAY.

So I say that THE way is to build it up from the grass roots and this means going so along the lines of BLEKKO.

Many future developments will evolve when we think correctly about this situation and drop the idea of a golden simple less work overhead answer other than the required hard work of inputting all the META attributes required.

Even then the results will be far from human intelligence.

Now whilst I’m on this subject: Human intelligence

Is there any evidence of this or in fact are we also subject to conditioned reflex thinking and automatic decisions,  that are similarly generated to operations of a search engine. I say we are because we are often asleep in our waking state.

Modern psychology has missed this aspect of human cognitive faculties: as most of the time we spend in REACTION to external events, we can be impulsive, emotional, even Pavlovian in our responses and our speech and not aware of decision processes.

For example: we are not always present and often ‘asleep’ driving to a location without remembering the journey; when we do habitual or repetitive tasks, these operations tend to turn off awareness, leaving an unconscious physical robot to take over, whilst the mind drifts through endless interlinked chatter, thought dreams, whilst externally to an observer we appear to be awake, and yet are we consciously ‘present’ from moment to moment…?



Psychology and brain researchers appear to miss study of these phenomena:

WHY?

Possibly because supposed ‘mystics’ have studied just such gaps in human consciousness – SELF conscious – The very thing that I claim that a computerised A.I. cannot achieve – we in fact most of the time do not posses this faculty: we possess it only intermittently!

For the most time we are not present!



 RenĂ© Magritte's Not to Be Reproduced




It is possible with a little mental effort to prove this phenomenon to yourself.



Look at an object in front of you, a tree, a cup; whatever, then BECOME aware of yourself looking at it. Your  ‘I’  your self -  looking SIMULTANIOUSLY at it… you should notice that you become fully aware of looking at it.

There you have it!

You the viewer perceive yourself - perceiving the object being - perceived.

This is conscious self awareness: and awareness/knowledge of SELF.

This may be what distinguishes mankind from all other species.

Its quite doubtful that a computational algorithm in a computer programme could do this sort of mental gymnastic… and its unnoticed most of the time for us humans as we often spend much of our days unknowing in an automatic dreamlike unconscious self none aware state .

This is not a Splodge idea; it was first introduced as an observation of mankind by George Gurdjieff.

 http://en.wikipedia.org/wiki/George_Gurdjieff


The problem was that Gurdjieff followers built on other ideas of his into a sort of cult, therefore ideas such as this one became tarnished an ignored by mainstream psychological thinkers.

But there is something significant in this… in understanding ourselves.

Tuesday, March 22, 2011

Social Media Advertising: the Final Frontier?

THE KILLER ADVERTISING PLATFORM?



Google has tried to buy Facebook several times but has always been thwarted by Mark Zuckerberg who owns the major part of Facebook shares. Further to the annoyance of Google is the fact that Facebook has allowed Microsoft to be a substantial investor: so no doubt the BING Search Engine may have a role to play in the future.


Currently no search engine is allowed into Facebook to catalogue its content.


http://en.wikipedia.org/wiki/Mark_Zuckerberg





SOCIAL MEDIA ADVERTISING - Versus – SEO: Search Engine Optimisation linked advertising.


Over 500 million users and growing: that’s Facebook


With Facebook it is possible to locate user groups with common interests that are friends of friends, who might pass on useful information: this equals word of mouth recommendations. This is the best that a product or a service can get, it’s the circle squared and the advertisers dream. Google can only estimate what users want from a search based upon input keyword descriptions which often misses the mark: link based ranking can falsely reach top positioned but not be what the searcher seeks. Mining the links was once input by people making these links somewhat meaningful, now its automated and low quality (e.g. MechanicalTurk).


Facebook is still the ‘next big thing’ and set to get bigger: with more integrated applications reaching special interest groups. If you want to reach all the people interested in a certain thing you can find them on Facebook. How do you locate and provide your service to all the people in a location that are interested in fly fishing, or people who like a certain genre of music? There all categorised on Facebook and can be reached with specifically targeted and quality tailored options to reach them, to interest them = less or no Spam. There is in fact no need to search outside of quality sites that give you the material – subject that your are seeking (see the Blekko video on the Spam, Spam, Spam article). So quality sites with integrated links to & fro in Facebook look a likely new killer Ap for it.


Also outlined in the last Splodge (Spam, Spam, and Spam) I mentioned that Google was putting up barriers to stop some of the existing methods that SEO advertisers utilise to favour their clients; by feeding keyword back-links to Google to get better positions for their clients.


A practice that has generated good fees for Google, but know it seems that Google has decided to forgo such revenues in favour of better more honest and accurate search returns for its users, is this all altruistic or is Google forced to do so because other ‘upstart’ search engines are focusing on Spam removal (Blekko) and thus could be a threat to the mighty Google?


Google may not even have to police some sites e.g. HubPages; because out of thier GoogleFear they have started to do so themselves with genuine and borderline Hubbers getting this message on their edit pages:
NEEDS REVISION


A red highlighted instruction appears on your HubPage edit and this is followed up with an automated email explanation of the infringement – it could be too many back-links (two maximum?) or content considered to be duplicating content on a none Hubpage site.
I tried investigating this and one way around it is to put the content first onto a HubPage and then a few days later, secondly onto another location, they then, don’t seem to mind! Ho Hum.


There may be other Social Network sites to come in the future but right now Facebook dominates.


Returning from my detour: all this well intentioned fuss may be just noise from Google because I can predict that savvy Internet advertisers will be moving in droves to Facebook once it is fully realised the power of that method over Google’s Spamful results.


Simply put Google has to guess by number crunching and matching keywords to ‘interests’ products and services, and its results are shamefully Spamful. Whereas Social Networks KNOW lots about their users and can match in an analogue fashion (people powered) and beside this can be placed very relevant adverts too potential clients.


Now ad this to the idea of Crowd Source Build Slash Tags (Blekko) and you get far less Spam, a better more accurate Internet, and more accurate advertising!


You heard it first from SPLODGE.

Monday, March 21, 2011

Spam, Spam, Spam

JUNK FEEDS – JUNK FOOD


Good legitimate content, advice and services are obscured by Spammers pushing their paid for advertising at us all, bloating the Internet with rubbish or worse: content that pretends to be professional advice!

These organisations have been making good incomes by charging their clients to get them to the first results page on Search Engines using  SEO techniques (Search Engine Optimization) and launching at Google and other result pages their clients URL’s due to a fee, rather than any real value.

The Internet has enough junk, and its time the detritus was removed, or at least blocked.
Why should ‘businesses profit from Spamming us all with junk feeds.

Blekko and Google have black balled content farms and certain aggregation sites and banned several of these and deleted links from many more legitimate content sites like Squidoo and HubPages.

slash the web!  What Blekko Search engine aims to do.

 
Blekko is a better way to search the web by using slashtags. slashtags search only the sites you want and cut out the spam sites. Use friends, experts, community or your own slashtags to slash in what you want and slash out what you don't.
Web search bill of rights from Blekko – Admirable…
1.
Search shall be open
2.
Search results shall involve people
3.
Ranking data shall not be kept secret
4.
Web data shall be readily available
5.
There is no one-size-fits-all for search
6.
Advanced search shall be accessible
7.
Search engine tools shall be open to all
8.
Search & community go hand-in-hand
9.
Spam does not belong in search results
10.
Privacy of searchers shall not be violated


Lets hope it works!

Content farms, low-quality duplicate content sites and keyword stuffed aggregators have be effected by the Google algorithm change introduced in February (2011) which filters out more Spam: for example HubPages dropped about four million hits. HubPages are much more legitimate in their aims, and such information can be obtained, as for the others its more difficult to assess their results: dropping I hope!




BUT… there’s always a but…


  

Camouflage tricks can be loaded into Squidoo, HubPages and other sites, with content masquerading and pretending to be one thing but loading up with every spelling and misspelling (TAGS) to generate links – camouflaged backlinks to assist hit values to AdSense ads.


GOOGLE Spam’s content farms:

On one hand this is good news for Internet users – fed up with hundreds of results that do not match their genuine need, but on the other hand lots of lost revenues for Google AdWords?

 

Spam site businesses that sell SEO content aggregation and gain from AdSense revenue could be in some trouble, possibly - eHow, eZineArticles.com etc but maybe Google will suffer too!

Google shooting itself in the foot?
Google’s enabled content farms and ‘backlink’ aggregators to be profitable in the first place with AdSense and made millions from the traffic generated, so adversely effecting these organisations and methods could be shooting Google’s own advertising revenue footprint.


The Blekko search engine and The War on Content Farms