Tagged: Artificial intelligence Toggle Comment Threads | Keyboard Shortcuts

  • Shareumentarian 12:37 pm on May 24, 2009 Permalink | Reply
    Tags: Abraham Maslow, arabesque, architect, architecture, , Artificial intelligence, , , ceramic, ceramics, Computer simulation, , , eramic tiles, , fractal geometry, green echo, interior design, interlock, japan, , karakusa, linked data, math, musashino art university, , patterns, persia, roots revival, Social Sciences, spice route, tile, tokolo, tokyo, tokyo institute of computer science, we are family, western individualism, wrapping fabrics   

    Bruce Mau Design 12 Strategies and Neemee.com 

    From Bruce Mau Design: “2.  Scale: the end of the discrete object. — Every object incorporates other objects just as it is itself incorporated by other systems.

    This refers to my earlier statement that “Scale is Meaning and Meaning is Scale”.  That statement translated into Culture Waves goes something like this:  “Wave Evidence defines a Human Truth, Human Truths define Human Behavior, and Human Behavior drives Wave Evidence” a bit circular I know but his statement is as well.  Ultimately, he is taking a Macro to Micro and then back again view of the interconnectedness of things. This is all fine and dandy until you get to the question of how this actually happens.  As a design principle its easy to apply this approach to a closed system and then rifle up and down the scales of things to define their connections but when you get to Human Behavior and the milieu of things that it touches it’s a different issue. This is where our Folksonomy and Ontology come into play.  Together they form a Folktology (I suspect that’s a word he will respect if he’s not using it already).

    Repeating Tiles

    The notion that the “end of the discrete object” is here needs a practice to support it, a language to define it and share it.  How do you understand that twelve hundred year old karakusa (in japanese) or arabesque patterns are anything more than innovative pattern making in a world of adaptive computer modeling.  How do you understand that this Green Echo behavioral evidence also has to do with Abraham Maslow’s notion of “Belonging” without making that inference yourself or having someone share it with you.  Besides the fact the Persian origin of these patterns are organic in nature, Green Echo sits Ontologically within Maslow’s Belonging category and the evidence is tagged with these terms, “green echo, belonging, design, deity complex”.  Take a look at the other thoughts within neemee also tagged with those ontological terms and you start to get a sense of the connectability of the CultureWaves Ontology.  The Tag Cloud associated with these thoughts then begins to provide a full picture of the very human concepts associated with these thoughts.  The Neemee/CW Folktology allows individuals to identify and tag discrete objects but then systematically incorporates them into larger connected structures of Human Behavior.

    Reblog this post [with Zemanta]
     
  • Shareumentarian 4:40 am on April 27, 2009 Permalink | Reply
    Tags: Artificial intelligence, , Information Retrieval, , SemanticWeb, social bookmarks, Topic Maps, Web 2.0, Yahoo   

    Folktologies Tool: Tag Mapper by Fuzzzy 

    screen-capture-691

    What is all the fuzzz about

    Why is it different from del.icio.us and other social bookmarking sites? Fuzzzy is not only social but also semantic. In fuzzzy Tags as well as their relations have more meaning. When tags are assigned a meaning using a standard like the ISO 13250 Topic Map then people as well as other computer systems can make use of the embedded knowledge in a more meaningful way. More on Topic Maps at Wikipedia. This way of categorising content is a middle way between the top-down monolithic taxonomy approach like the Yahoo directory and the more recent social tagging (folksonomy) approaches. It is also more collaborative and will help to build a consensus language that can improve communication and support knowledge sharing

    screen-capture-70

    Messy or just fuzzzy?

    We call the mesh of tags a simple fuzzy common ontology (folktology) or a semantic tag cloud. Our folktology has a set of interrelated tags where each tag has a list of synonyms or aliases, a short description and relations to other tags. This makes the semantic tags suitable for categorization and can be used by various applications. The name fuzzzy comes from fuzzy logic in Artificial Intelligence. On fuzzzy.com we do not use the simple plain text tags which are common to many other web2.0 sites on the other side we do neighter adhere to more formal ontologies or description logic languages.

    tag-mapper1

    Instead we employ a pragmatic middle way approach where we use Topic Maps and a simple model where users create a semantic network of tags that may overlap end even look a bit messy. The word fuzzy is here used to describe how tags can be inaccurate. Human knowledge and language are fuzzy, so creating an accurate tag set that makes sense to everyone of the world is impossible. Instead we let people create semantic tags as they like in a bottom up fashion. The resulting tag set will be a bit fuzzy but tags still have much more meaning than plain text tags. fuzzzy tags will lower the ammount of noise and there is a much greater chance you will be able to make sense of the tags and you might even learn something in the process of browsing tags.

    How is it different from other social bookmarking services?

    fuzzzy is not just social but also semantic, collaborative and to an extent democratic. Users collaborate on creating tags, linking resources and connecting tags and knowledge resources together. Tags can also have relations between them. The underlying semantic Topic Maps system enables global distributed tagging. This makes fuzzzy unique. fuzzzy also stands out from the crowd in that it is the first bookmarking site intended strictly for learning resources.

    fuzzzy future

    The system is developed in tandem with academic research in the fields of Information Retrieval, HCI, Semantic Web, Knowledge Organization, Knowledge Management and Knowledge Federation. We are continuously developing the system and we plan to do so the years to come.

    Our research is in stage one of three. Stage one consist of getting the portal up and running and establishing a community. Stage two will consist of tuning, developing democratic features and evolving the system in accord with the community. Stage three is all about hooking on the Semantic Web, integrating knowledge and making the knowledge within fuzzzy available to other Semantic services. The system platform developed and knowledge gained through this project will be reused on the planned Omniscious project.The aim of the Omniscious project is to provide an information infrastructure for collective wisdom and for collaboratively solving complex global issues, enable the creation and sharing of holistic knowledge.

    Reblog this post [with Zemanta]
     
    • Alex 11:42 pm on April 27, 2009 Permalink

      Great writeup. Can people also look at or share the Topic Map fragments (XTM?) of a particular tag or cloud?

    • Robert Barta 12:28 am on April 28, 2009 Permalink

      Makes all sense to me. I have the same question as Alex: is there a way to exchange something with the service?

    • Roy Lachica 8:25 pm on April 29, 2009 Permalink

      The blog post above is a copy of the about page on fuzzzy.com. After the relase of version 2.2 of fuzzzy the core engine have been totally rewritten. Full XTM2.0 fragment support will be available in a month or two.

  • Shareumentarian 3:52 pm on January 26, 2009 Permalink | Reply
    Tags: Artificial intelligence, Controlled Vocabulary, Culturewaves Controlled Vocabulary, , , , , , , , Mass Collaboration, , , , Protégé, , , ,   

    What are you thinking? 

    What are you thinking?  Describe that to me so I can understand and maybe we can collaborate on what ‘we’ are thinking.  This is the charge of any collaborative environment; a download, a sharing, a communication of shared ideas so that commonalities can be worked on, mulled over, and then mashed up so the third thing, the elixir of communities, can be made: RelevanceCollaboration is an oft-used word these days but as anyone who has participated in a successful collaboration knows, a facilitator is key for any sense to be made of the myriad of collaborative contributions.  From an Ontological perspective, this is no less true of artistic collaborations than it is of Insight-based collaborations; and often these are exactly the same.

    Neemee.com is a social bookmarking tool based in an ontological methodology of structuring its bookmarks, or captures, through the Controlled Vocabulary of Culturewaves and an Ontology of Four Life Categories and Maslow’s Hierarchy of Needs.  The nature of Neemee.com’s real-time capture device, called the Drag-and-Drop box, allows for the bookmarks stored

    Neemee.com's Drag and Drop Box
    Neemee.com’s Drag and Drop Box

    within Neemee.com to be compiled from multiple sources rather than one historical link out on the web.  The nature of multiple sources referencing one aspect of a Controlled Vocabulary is more akin to a thought, thus Neemee.com’s bookmarks are called Thoughts.  When someone asks you what you think about the current economy you don’t just reference the WSJ article you read this morning and call it a day.  Your response to that question is going to involve that WSJ article, the decision you made during lunch for one or separate checks, and the amount of tension you feel over the liquidity or non-liquidity of your finances right now.  And so we come back to the beginning of this post: Sharing.

    Contributors to Neemee.com share a vocabulary called Culturewaves.  A proprietary insights tool designed to identify emerging behavioral trends; Culturewaves sifts behavioral evidence, objects and events, through a meaning based Ontology of Four Life Categories and Maslow’s Hierarchy of Needs, and Culturewaves’ Controlled Vocabulary of Wave names and associated Human Truths. The foundation of Culturewaves is a flexible meaning based categorization methodology that can absorb changes over time.  The two together, Culturewaves’ controlled vocabulary and the Neemee.com Ontology, provide an input into Neemee.com that is both intuitive and scientific, artistic and research-based, individual and collaborative.  It is less a calcified Ontology than it is a Juxtaposition of Relevance.

    All of this is happening in real time.  Neemee.com members are submitting thoughts at all hours of the day and night and the tag cloud, the crystallization and display of all members’ tags, grows by the minute.  It is, for all intents and purposes, Mass Collaboration.  The collaboration exists, in its most apparent form, through an Ontology of Life Categories and Maslow’s hierarchy, and Culturewaves Controlled Vocabulary.  This is where an Ontology Director steps in as a facilitator.

    What are you thinking?  Asked collectively, this question will illicit a mass groan for its ambiguity or a myriad of specific answers for its opportunity.  It is the job of the Ontology Director to sort all of these out.  The existing tag cloud in Neemee.com contains the answers to this question but in an amalgamated form.  With the Neemee.com Ontology, Culturewaves Controlled Vocabulary, and the multitude of tags, or Folksonomy, attached to each thought by Neemee.com users, the hierarchical nature of an edited ontology becomes useful.  Macro terms such as Design, Entertainment, Technology, and Well Being (Life Categories) give context to Maslow’s Hierarchy.  From the bottom up, Culturewaves terms like Pop up Lifestyle and Green Echo bring definition to the various conceptual realities contained within, for this example, the Belonging level of Maslow’s Hierarchy.  Added to this Linguistic Structure of Meaning, are the purely descriptive folksomic tags that users apply to their thoughts.

    Given the ‘Crowd Sourced’ nature of the Folksomic tags, the more structured ‘Ontology’ and ‘Controlled Vocabulary’ becomes flexible in its input.  The Folksomic tags inform, from a Human Behavior perspective, the diversity of meanings and relevancies that the Ontology and Controlled Vocabulary seek to generalize and make more uniform.  The dialogue between the three creates a kind of Human Algorithm of Relevancy, or Relevantology.  The Ontology Director to applies the language of tags surrounding the Human Behavioral Evidence, or Thoughts, to the implied meaning of the Ontology and Controlled Vocabulary to create a common landscape of collaboration where thoughts from a myriad of Neemee users can be filtered and evaluated regardless of their seeming disparity.  The Architecture of the Semantic Structures within Neemee.com allows the Ontology Director to apply differences of Language and Languages toward the same World Thought Bank without any loss of Meaning or Relevance. Through Ontology Editors such as Protégé, different terminologies and cultural languages can be aligned to maintain the significance of their usages.  The World Thought Bank, through this alignment, becomes a Multi-Cultural and Cross Business platform for Mass Collaborative Insights.

    There are existent Semantic Ontologies and Communities that can be leveraged to support the Semantic Architectures created within Neemee.com.  Through Ontology Matching, correspondences between semantically related Ontologies can be identified and utilized to bring wider Relevance to the World Thought Bank.  To align Neemee.com’s Semantic Architecture to Ontologies that are being, and have been, created utilizes Neemee.com’s real-time insights to bring communities together that are actively working on increasing the Relevance and Meaning of the Semantic Web.

    What are you seeing? The Formalization of Insights makes them ultimately transferable and shareable.  Through ‘Tag Threads’, ‘Semantic Evidence Crowds’, ‘Relevance Juxtapositions’, Etc., the insights uncovered through Neemee.com’s Semantic architecture becomes an interactive road-map through which clients and Neemee.com users can navigate their way to greater relevance.  Neemee.com users can find where their thoughts intersect with others and find greater inspiration through multiple perspectives and relevant juxtapositions.  Neemee.com clients can take these relevant juxtapositions and apply them to their business through the specificity of the Folksonomy, related Neemee.com thoughts, and generalities fleshed out by Neemee.com’s Ontology and Culturewaves’ Controlled Vocabulary.

    Reblog this post [with Zemanta]
     
  • Shareumentarian 3:50 pm on January 8, 2009 Permalink | Reply
    Tags: Artificial intelligence, , , , , , , Markup Languages, , , , thinking, thought   

    Curiosity Evoked 

    The thoughts flowing through Neemee are enticing as individual calcifications of one’s journey through the web.  As enticing as they are, even exciting, there is always the question of where they came from.  The sources are there for sure but that’s not the origin I’m talking about. The curiosity evoked is not so much the original location of the thought but the origin of thought behind the thought; what was the thinking.  It’s impossible to get that perspective without some juxtaposition, without some counterpoint, without some connection.

    Neemee Collections begin to do that, collect multiple thoughts with some intended purpose, but ultimately the connections are implied rather than explicit.  To bring clarity to the multiple channels of thinking going on inside Neemee, rather in-depth thinking I might add, connections need to be made, visual connections.  The classification of thoughts within Neemee, the Ontological organization, occurs at the level of tags initially (tags have a greater purpose than mere identification) but through the convergence of the Ontological tags (Life categories, Maslow categories, Wave names) thoughts begin to cluster around common concepts: Technology, or Belonging, or Pop Up Lifestyle, or all three of these together.  It is this clustering of concepts that begins to create neighborhoods of conceptual meaning within Neemee.  Given the three Ontological tags, any thought can be grouped with all of the other similarly tagged thoughts.  This is at the neighborhood or conceptual level, now let’s look at the street and address

    The folksonomy tags, the words applied to a thought to define it, begin to zero in on the street and address of a Neemee thought. These folksomic tags:  tv, television, mobile, cell phone, verizon, mp3, gps, networks, att, broadcast, cellular carrier, qualcomm from the thought “Broadcasts to mobile devices to start in 22 cities” define the neighbors and inhabitants of the thought, but beyond that you have no idea where you are in the metropolis of Neemee meaning.  As neighbors do, the folksomic tags reach out to touch each other.  Clicking on “att” will bring you to all the other thoughts similarly tagged.  A complex web indeed, but to bring the meaning of all this together though, we need a way to visualize their connections.  This brings me to the outside reference of this blog post.

    Yusef Hassan Montero and Víctor Herrero Solana of the University of Granada, Spain have posted a visualization of delicious.com bookmarks that gives a tangible, and clickable, map of related posts.  I’m just going to give the link and let you check it out for yourself.  Once you’ve experienced it, I feel confident that you’ll see the value and power of participating in Neemee’s Folktology (Ontology+Folksonomy): Visualizing Social Indexing Semantics.

    Reblog this post [with Zemanta]
     
    • nobboble 5:15 am on January 11, 2009 Permalink

      stuat…this is good stuff and needs to be on the neemee site… i’m passing it along to leslie and caitlin to set up….with the formal heading of you as the director of ontology… let’s get publishing…

  • Shareumentarian 8:52 pm on January 5, 2009 Permalink | Reply
    Tags: Artificial intelligence, , , , , MusicBrainz, , , ,   

    Zemanta: Collaborative Thought Through Borrowed Relevance 

    Image representing Zemanta as depicted in Crun...
    Image via CrunchBase

    Does that make sense to you? It’s a question we ask those we communicate with all the time. Children ask it because parents are the most illogical people above three feet tall, your co-workers ask it because they don’t want to be the only ones carrying the ball, innovators ask it (or at least they should) because if laymen can’t make sense of their innovations then implementation becomes an issue. In light of all that, making sense of something is the primary job of language itself. In the face of Mass Collaboration, language is strained to do its most delicate job of describing nuance; in fact, it does its best just to keep up with broad generalities at such a massive scale as, say, all of the blogs in the Blogosphere.

    Enter the Semantic Web and the myriad of tools that have sprung up to help language do its job, in this case, Zemanta.  In their words:

    “We analyze your post through our proprietary natural language processing and semantic algorithms, and statistically compare its contextual framework to our preindexed database of content.

    We are using a combination of machine learning techniques and end-user input from our widget users, that enables us to train the engine and constantly improve the recommendations.”

    Zemanta distills meaning from language on the Web using multiple methodologies that can be grouped in two basic categories: HI (Human Intelligence) and AI (Artificial Intelligence). Their Natural Language Search and Semantic Algorithms digitally identify common terms and assign them to their appropriate groups such as Names, Places, Companies, Date, Actions, and Emotions. Zemanta has another source of collaborative input that is constantly updated and more flexible than pre-classified language terms or mathematical algorithms.

    Zemanta has, quite smartly, joined the world of user widgets and proliferated their Semantic integration technology around the web. From millions of bloggers using Zemanta’s Semantic technology to identify tags, images, and related links for their posts, Zemanta receives a truly Crowd Sourced bank of linked Web connections. There is an API for web developers to integrate into their site as well.

    “Zemanta fully supports the Linking Open Data initiative. It is the first API that returns disambiguated entities linked to dbPedia, Freebase, MusicBrainz, and Semantic Crunchbase. The data can be returned in the standard format of Semantic web – RDF. It is an ideal gateway from unstructured web to semantic web. This represents a major step ahead for efforts to connect the Web into a semantic web of objects.”

    Neemee takes this philosophy of defining and capturing only the relevant pieces of Web pages to the world of Social Bookmarking. Not only does Neemee rely on HI to build user thoughts, a single thought can be from multiple sources on the web so the “Linking Open Data initiative” becomes further integrated and folded back onto itself. Neemee’s Drag and Drop capture device allows Humans to piece together a complete thought from all of the many voices on the Web. Combined with Semantically Integrated web sites like Zemanta, Neemee becomes a powerful tool to combine HI with all of the sophisticated AI engines sifting and sorting the web.

    Reblog this post [with Zemanta]
     
    • Andraz Tori 10:40 pm on January 8, 2009 Permalink

      Thanks for sharing this interesting perspective on our technology.

      We are definitely hoping others (Naamee included) will use our API to provide even better experience to their users. Some sites like Faviki and Kakuteru are already doing that. For example you could start with few paragraphs of text and get back images that can be starting points for your thoughts.

      Have fun!
      Andraz Tori, CTO at Zemanta

    • kebadydex 3:53 am on December 10, 2009 Permalink

      Lots of of folks write about this matter but you said really true words.

c
Compose new post
j
Next post/Next comment
k
Previous post/Previous comment
r
Reply
e
Edit
o
Show/Hide comments
t
Go to top
l
Go to login
h
Show/Hide help
shift + esc
Cancel