Project Overview

                Draft - Do Not Quote Or Distribute

Page Summary: Commonsense knowledge is widely considered to be one of the most difficult issues in AI. In this theoretical approach, we apply a paradigmatic relationship layer to a lexicon in which the data components are related to form word-topics. The product resembles a thesaurus. It is a schema relaying word-topics (i.e., sense-units) to natural language programs that aim to analyze text requiring high levels of commonsense reasoning. The schema doesn’t provide the reasoning. Rather, it provides the word-topics that act like motivated-dynamical-contexts or orientational gadgets. In this article, we introduce the gadgets, knowledgebase and schema. The project includes a wiki, source code, various acquisitions & interfaces. A Word-Sensibility Model is introduced as a concept basic to the project.

The Quadranym Model of Word-Sensibility (Q): An Ecological Systems Perspective On Word-Level Concepts & Contextual Unitizations – Non-Mental-Representation Representation Design Before Define Approach.

The application aims to assist access to data in dictionaries as pertaining to commonsense situations in text; a method concerned with computational lexicography and ontology for AI. It involves a meta-lexicography as it aims to deal with accessing word sense using lexical paradigms to search words and simulate the human ability to sense dynamic relations between signs.

In the Q, the signifier requires two different systems of context to be signified. Q analysis makes a distinction between the situational context (world conditions) and the dynamical context (organismic responsiveness). The Q is about an ecological system of agency for accessing knowledge.

The project is about developing an abstract representation for how human responsiveness responds to written text. Some inspiration comes from the concept of experiential traces as found in the embodied language processing hypothesis. The goal is textual analysis to assess word sense by keying on human responsiveness (i.e., the ability to act quickly and positively to situations in the world). It is about commonsense grounding and metaphoric structure. Ideally, it aims to help assess dynamic word relations of the type found in mythic stories. Our current goal is to build an ontology and API pertaining to a lexical layer of word-sensibility that sits between a lexicon like WordNet and a knowledge graph like ConceptNet. Central to the idea is intersubjectivity, specifically how people share orientation and focus and how this dynamic can be extended to machines.

The Dynamical Context

Unpacking words with Q analysis begins with a distinction in the contextualizing of word sense. A Situational Context is the communicative ability to present or understand the objective circumstances in which an event occurs and will sometimes include the appropriate behaviors associated with it. We introduce the idea of a Dynamical Context which is something different and can be summarized as follows:

  • Dynamical Context: a situation resonates with a preexisting psychology, a predetermined expectation for behavior within that situation, and produces a synergy response, reshaped for the moment.
  • Dynamical Contextual Systems: characterized by the potential for multiple dynamic areas and interactions between them.

We introduce conceptual constructs called quadranyms and polynyms. The terms when pertaining to system processes form a matrix that represents a general and overarching cognitive framework for a Dynamical Contextual approach for an ontology of commonsense. Humans are wired to connect. The focus is on the dynamicity of a word’s specificity to relate word sense.

 Introducing Quadranyms & Polynyms
Featuring a Wiki & Acquisition Database

We introduce a method to acquire sets of conceptual relations in a format of dimensions. We call it nymology, a practice of collecting and organizing idea sets: each a set of terms representing a certain number of dimensions used to strategically unfold, frame or simplify a concept. Each set functions as either part, step or type. A dimension number is called a polynym i.e., mononym, duonym, trionym, tetranym, pentanym and so on. Entries include, area, source, URL and logged by. A collection resembles a thesaurus and can be used to help populate knowledgebase systems and enhance queries.

What is a Polynym?
We use Polynym to describe idea sets (as part, step or type), such as:

  • 3 parts × Freud’s psyche = [ Id, Ego, Superego ]
  • 7 types × Deadly sins = [ Wrath, Greed, Sloth, Pride, Lust, Envy, Gluttony ]
  • 5 steps × Grief = [ denial, anger, bargaining, depression, acceptance ]

There is a difference between artifactual polynyms based on long term study (accepted or rejected) and system processes representing polynym constructs assembled for the moment, to contextualize a situation.

  • Polynym idea sets are strategic ways to think about a topic. Strategies pertain to the Situational Context. That is, what is true or false. It is a  deliberative process in that a topic is about what the condition is.
  • Polynym idea sets exist in virtually every discipline and are abundant. Polynyms are also important to a system’s basic process. They are about the truth conditions of the real world, including attitudes and behaviors.

What is a Quadranym?
We define Quadranym as a four-part conceptual construct using a dual-axis Mode-State model. It operates as a virtual unit of orientation & constraint.

Expansion-Reduction (ER-mode) and Objective-Subjective (OS-state).

  • Quadranym idea sets are about framing a response to a topic. Responses pertain to the Dynamical Context. That is, how one copes. It is on the heuristic level in that a topic is about how to respond to a condition.
E = expand: open
  O = object: barrier
  N = Topic Name: door
S = subject: passage
  R = reduce: close

A responsiveness for door is rendered, a dynamic framework for specified terms; a constraining unit specifying a virtual motivated-dynamical-context.

The quadranym (word-topic) model is inspired by embodied language processing involving motor and sensory perception and experiential traces.

Word-Topics (Micro-Topics); help organize, characterize or summarize lexical information. They’re pre-textual micro-unit-renderings of context used to anchor focus given to a sentence and to contextual development.

Quadranym Representation Introduction:

The quadranym square divides into left and right hemispheres, top and bottom levels and diagonally related on mode & state axes.


The typical configuration is as hemispheres as represented as follows:

  • Left Side (Superset): Active_Potential(active_actual)
  • Right Side (Subset): Passive_Actual(passive_potential).
  • Superset = Active Sense: more experience necessary (find potential)
  • Subset = Passive Sense: no more experience necessary (potential ready)

Active refers to the unit using power. Passive refers to the power being used. Active sense initiates a word-topics function and argument. Passive sense is when the function and argument is configured for the situation. Actual state is the source. Potential state is the target. For instance, in the next example, the active sense of space is void. The passive sense of void is between. Between is whatever the situation is. If it is the walls of a room or a passage between rooms, then it is those specific things. Void (or emptiness) remains active no matter what the situation is. The active sense is the question, the passive sense is the answer. The quadranym does not necessarily give the answers, rather it aligns its active sense with the text it is analyzing to make sense of the situation as a potential of the real world.

Terms are given these attributes and cluster in the sets as follows:

  • The superset is FOR the source condition terms (FOR all terms).
  • The subset is FOR the target variable terms (FOR some terms).


Each dimension holds its own categorical set. For example, open holds over, inclusive, infinite, out and all as some factors that unpack given the context.

During the analysis of a text, all of the text begin as the superset and then gets filtered into subsets. The process separates constants from variables

Cluster: {…}

  • Open{…)(passage{…}) ⊇ Close{…}(Barrier{…})

Each quadrant (Meta-role) is essentially an adjective. Anything in a particular quadranym category is described by that quadrant. For instance, barrier is FOR something as general as variance and something as specific as security. Any term in a quadranym category can be thingified. For example, run, ran, running, can all be run, ran, running, things. And for the most part, any quadranym category can be FOR them. As much as it might not make sense, barrier can be FOR running. Consider the sentence, “We need to stop people from running over the new grass”. The system aims for a synergy between it’s categories and some text (a potential for any actual).

  • The process is not about identifying a situation but its potential dynamics.

Consider the utility of door as described in the following situation…

Situational  Context

Text: “Close the door because it’s cold outside.”

Dynamical Context

Topic Factors: {time, space, door, valence, temperature}

Meta-Dimensional Roles: {now, space ,open, door, close, temp, affect}.

Content_Roles: {<Close, the_door, because, it’s_cold, outside>}

There are many ways in this model to parse text into units below. However, there are relations ready made in the script. The Content Roles represent the text and Meta-Dimensional roles represent word-topic categories. It’s best to think of the M-Roles as being FOR something is the text. Notice how close is in the future of now. The affect affect directs the motivation

space: Infinite{invariant)(void{outside}) ⊇ Finite{variant}(between{temp})

affect: Positive{warm)(affect{because}) ⊇ Negative{cold}(situation{change})

temp: Cold{out}(optimize{Close}) ⊇ Warm{in}(condition{cold})

open: Open{cold)(passage{the_door}) ⊇ Close{warm}(barrier{event})

now: Future{close}(present{it’s_cold}) ⊇ Past{open}(event{door})

The example shows how quadranyms can configure statistical models. In this realm, door is configured to assess optimization in a climate context.


100% closed is better than 90% closed. 90% closed is better than 50% closed. Temporal and spatial factors configure together to optimize desired conditions; procedure = less open & less time open. If door remains open desired temperature will be reduced. Because door is a subset of the spatial domain it is nested in spatial dimensions. This form is illustrated in any matrix table of nested topics. The point being, the variability in climate is dependent on the division of spatial modes, all_out & some_in. All space is out temperature non-discrete. In the door realm, any variance is dependent on barrier, i.e.,  close_in is the independent variable that influences open_out temperature. All Quadranyms can conform in this way in some manner.

Finally, the idea is for machines to have the ability to abstract the human experience of contextual responsiveness, building that abstraction up from the word-level, and in this way machines can have some contextual expectation to make better sense of the things that people typically know.

Habit is the Social Glue

Next: System Overview



© 2000-2014 All Rights Reserved

Draft – Work In Progress – Do Not Quote Or Distribute