System Overview

The Complete System Overview. System Summary Version: Home Page

Context isn’t just the surrounding circumstances, because it includes and interacts with the subject that is surrounded, and the agent that tries to comprehend it all. ― Andrew Hinton, (2014). Understanding Context

This page is waiting for updates.  

Draft – Work In Progress. 

Visit the About page for the project’s motivation, scope and goals.  We are in the exploratory stage and look forward to any feedback. Please Contact.

1) Word-Sensibility is About Orienting to Communications:

Word-Sensibility is a theoretical approach that can be interpreted for a wide span of applications: from  API services for NLP to metaphor-analysis.

The theoretical ideas are most concerned with the non-linguistic aspects of using language, aspects that ground the responsiveness of communication.

2) Normative Responses for Artificial Intelligence:

Word-Sensibility is proposed as a model for context analysis. Data training involves a database with its features tagged to identify apt reference frames. 

  • Context analysis is about how text is contextualized. The word-sensibility approach pivots on mixing and matching an agent’s normative responses. 

Consider the sentence: Let’s move the couch over there.

  • Content: {<let’s, move, the_couch, over_there>}.

General reference frames aptly apply, such as: agent, energy, time and space.

Matching: Agent(let’s), Energy(move), Time(the_couch), Space(over_there)

(Note: The matching-scope is a variable that varies the inclusion radius. Content can be added to one or divided or duplicated to any number of reference frames.)

Reference frames are a kind of bias that allow assumptions about topics.

3) Relevant Responses Require General Responses to Anchor On:

The relevant response is about why people move furniture around. The general response is about intuitive grounding and orientation. The aim is to create Systems of Orientation by layering frames  •  scripts form on layers.

  • Systems of orientation combine, modify and layer reference frames (units).

Reference frames anchor on word states that target topic variables:

  • If agent: then self targets goal to find ‘agent’ topic variables. 
  • If time: then present targets event to find ‘time’ topic variables.
  • If space: then void targets between to find ‘space’ topic variables.
  • If energy: then motion targets matter to find ‘energy’ topic variables.

The system works to engage states e.g., agents in a temporal sequence, in a space, using energy. The sentential content is parsed to anchors and targets.

  • Words (text data) cluster in reference frames becoming target variables.

(Note: Content is (aptly) predicated by the reference frame it is in. Keep in mind that there are many ways to link grammatical structure to the reference frames.)

4) Source and Target Conditions: (Semantic Relations of Word-Sensibility )

The anchor states (i.e., source condition) are self, present, void and motion. The target states (i.e., target condition) are goal, event, between and matter.

  • Anchors are unmeasured so anchors require targets to find measure.

In the space frame, the anchor void targets between to find spatial measure.

  • Target states become anchor states as apt frames become more relevant.

Note:  Reference Frames take aim at condition variables based on what’s given. In the spatial frame, void is what’s given for any topic, and takes aim at the variables belonging to the term between ― such as, objects, regions, solid separation, spatial separation. Void is the source affordance for spatial context. Between is the relevant topic of space and the measure of the source condition.

Between can become a new anchor and target more relevant measures such as, house, room, person or couch (each has a mean). This is essentially a script.

  • [void ➝ between]:[between ➝ spatial separation(region➝ object(?…couch))]

(Note: Spatial frames such as, distance, direction, container, door can also apply.)

[def_task ’match reference frames to sentence’  

   :find [’pull together a set [? reference frame]’] 

     :result [[contenders ’x=frame, x=frame, x=frame, x=frame’]] 

        [Task-Find[Task ’if [?] add reference [?], otherwise if [? ]  [? ]’]] 

            [Task-Find [task [? system] agent #1, time #2, space #3, energy #4’]]]

5)  Orientation, Nested Layers and Viewpoints:

Initiate orientation: the general topics orient to the relevant topic.

Orienting for relevant time-space:{<the_couch, over_there>}

  • [present ➝ event]:[event ➝ sequence(start ➝ engage(?…the_couch))]
  • [void ➝ between]:[between ➝ spatial separation(region(?…over_there))]

Orient for relevant topic: {<move, the_couch>}

  • The relevant topic of the situation: action(s) {move} and object(s) {couch}.
  • Time and space reference frames orient for relevant temporal-spatial sense.
  • Anchors: If time then move<isIn>present. If space then move<isIn>void.
  • Targets: If time then couch<isA>event. If space then couch<isA>between.

(Note: Symbols <isA>, <isIn>, <use>, <find> or <x> are links for state variables.)

Relevant topic: (Nested Layers for time and space)

  • Time: anchor = Present(move) ➝ target = Event(couch)
  • Space: anchor = Void(move) ➝ target = Between(couch).

Orient for relevant viewpoint: {<let’s, move>}

  • Agents(let’s) <use> Energy(move).

Orient agent to energy viewpoint: 

  • If motion (i.e., actual-sense) then <find> matter (i.e., potential-sense).

From Agent To Energy: (Nested layers are vertical. Scripts are horizontal.)

  1. From: Agent: From Self(let’s) To Goal(move).
  2. Orient:Time: From Present(move) To Event(couch)
  3. Orient:Space: From Void(couch) To Between(there)
  4. To: Energy: From Motion(there) To Matter(couch).

( Note: From-To designates the scope radius for the scripts and layers.)

A hierarchal system for orienting a relevant viewpoint takes shape.

  •  Next task: align the relevant viewpoint to the situational dynamics.

6) Word-Sensibility Proposes Two Complimentary Systems of Context:

The Situational Context is about moving (v) the couch (n-ob).  We introduce the Dynamical Context*. It’s about the responsive alignment to a situation.

  • Content collected in each reference frame has a specific Dynamical Context.
The Dynamical Context Couples with The Situational Context.

States & Modes: 

Images illustrate mode-variables. left: active/passive, right: size/weight.

  • Modes: dependent y and independent x 

Energy States Find Modes of Measure for Agent Viewpoint: 

  1.  General: self+motion = yActive(move) ➝ goal+matter = xPassive(couch)
  2. Relevant: self+motion = ySize(move) ➝ goal+matter = xWeight(couch)

Reference Frames Calibrate to a Situation (Situation + Mode-Set).

  • Calibrating matter (i.e. target) affects the sense of motion (i.e. anchor)

7) Quadranym Systems:

Example: (Energy).

Phrasal Template 1: (Mode Controls)

  1. x of Energy is actual-measure that controls the potential motion.
  2. y of Energy is any potential-measure for motion dependent on x.

Energy Quadranym: (Semantic Set)

  • Active, Passive, Matter, Motion

Quadranym Representation: (Semantic Resource Variable)

  • (∀x)  energy(x) ⟹ [Active(motion) ⊇ Passive(matter)(x)] 

Phrasal Template 2: (Semantic Procedure of Unit)

  1. General (anchor): For the motion of energy;
  2. Relevant (target): Active is Dependent on Passive to Find matter

(Note: Any quadranym phrasal template can apply to any quadranym.)

Energy + Content: (Semantic Alignment)

[Active(motion:(y=Let’s, move, over_there)) ⊇ Passive(matter:(x=the_couch))]

8) Alignment: (Virtual alignment with the world.)

To anchor the energy topic to any situation, active is dependent on passive. A more relevant set of mode-variables (size & weight) target the X(?…couch).

  • Agent’s viewpoint of motion = weight effects potential, independent of size.

Database representation: script and nested layers with mode variables.

  1. Script: energy:[Y(motion) ⊇ X(matter)]<find>[Y(matter) ⊇ X(?…])
  2. Layers: agent:[Y(self) ⊇ X(goal)]<find>[Y(motion) ⊇ X(matter])

Time and space with mode variables. Scripts for the current example:

  1. Time: [Y(present) ⊇ X(event)]<find>[Y(event) ⊇ X(?…the_couch])
  2. Space: [Y(void) ⊇ X(between)]<find>[Y(between) ⊇ X(?…over_there])

Alignment depends on situational variables e.g., couch, vase or mountain.

  • This is a dynamic agent situation: Time_Space_Energy to target the matter.
  • Utilization: a regression analysis based on similar situations can be utilized.

Two distinct systems of context adds stability and versatility for analyses.

  1. The orienting above can be used for other situations e.g., reaching or passing.
  2. The system orients for prediction-conditions and counterfactual-conditions.
  3. A system of orientation can be repurposed to ground metaphoric expression.

(Note: Script and hierarchical structures imply linear and logarithmic algorithms. Scripts can be thought of in Linear ways while hierarchies are more Logarithmic.)

9) Summing-up:

Reference frames nest into various hierarchical layers (i.e., from general to relevant). These nested layers form virtual systems of responsiveness. The approach aims to assist with hard Artificial General Intelligence (AGI) tasks.

Such as:

Q Systems & Models:

  1. Quadranym Database: semantic sets used to populate reference frames.
  2. Quadranym (is a set) e.g., (∀x)  time(x) ⟹ [Future(present) ⊇ Past(event)(x)].
  3. A reference frame models an orientation, or a layer of a system of orientation.
  4. Quadranym Ontology: specified models of orientation and dynamic relations.

Q system reference frames can be used with deep learning models, that utilize word embedding methods,  to orient the word associations learned.

  • Initial development requires attested reference frames and apt training data.
  • Developing a Q system requires a Neuro-Symbolic (hybrid AI) approach.

Visit the About page for the project’s motivation, scope and goals.  We are in the exploratory stage and look forward to any feedback. Please Contact.

*(Note: The dynamic context communicates the changes in a situation while the dynamical context communicates a system’s response to changes in a situation.)