Skip to Content

Changes for page Automation

Last modified by Robert Schaub on 2025/12/22 14:32
From version 1.3
edited by Robert Schaub
on 2025/12/22 14:32
Change comment: Renamed back-links.
To version 1.2
edited by Robert Schaub
on 2025/12/22 14:32
Change comment: Update document after refactoring.

Summary

Details

Page properties
Content
... ... @@ -1,31 +1,21 @@
1 1  = Automation =
2 -
3 3  **How FactHarbor scales through automated claim evaluation.**
4 -
5 5  == 1. Automation Philosophy ==
6 -
7 7  FactHarbor is **automation-first**: AKEL (AI Knowledge Extraction Layer) makes all content decisions. Humans monitor system performance and improve algorithms.
8 8  **Why automation:**
9 -
10 10  * **Scale**: Can process millions of claims
11 11  * **Consistency**: Same evaluation criteria applied uniformly
12 12  * **Transparency**: Algorithms are auditable
13 13  * **Speed**: Results in <20 seconds typically
14 14  See [[Automation Philosophy>>Test.FactHarbor.Organisation.Automation-Philosophy]] for detailed principles.
15 -
16 16  == 2. Claim Processing Flow ==
17 -
18 18  === 2.1 User Submits Claim ===
19 -
20 20  * User provides claim text + source URLs
21 21  * System validates format
22 22  * Assigns processing ID
23 23  * Queues for AKEL processing
24 -
25 25  === 2.2 AKEL Processing ===
26 -
27 27  **AKEL automatically:**
28 -
29 29  1. Parses claim into testable components
30 30  2. Extracts evidence from sources
31 31  3. Scores source credibility
... ... @@ -35,12 +35,9 @@
35 35  7. Publishes result
36 36  **Processing time**: Typically <20 seconds
37 37  **No human approval required** - publication is automatic
38 -
39 39  === 2.3 Publication States ===
40 -
41 41  **Processing**: AKEL working on claim (not visible to public)
42 42  **Published**: AKEL completed evaluation (public)
43 -
44 44  * Verdict displayed with confidence score
45 45  * Evidence and sources shown
46 46  * Risk tier indicated
... ... @@ -85,45 +85,33 @@
85 85  * **Human review queue** for low-confidence verdicts
86 86  * **Independent claim processing** - errors in one claim don't cascade to others == 3. Risk Tiers ==
87 87  Risk tiers classify claims by potential impact and guide audit sampling rates.
88 -
89 89  === 3.1 Tier A (High Risk) ===
90 -
91 91  **Domains**: Medical, legal, elections, safety, security
92 92  **Characteristics**:
93 -
94 94  * High potential for harm if incorrect
95 95  * Complex specialized knowledge required
96 96  * Often subject to regulation
97 97  **Publication**: AKEL publishes automatically with prominent risk warning
98 98  **Audit rate**: Higher sampling recommended
99 -
100 100  === 3.2 Tier B (Medium Risk) ===
101 -
102 102  **Domains**: Complex policy, science, causality claims
103 103  **Characteristics**:
104 -
105 105  * Moderate potential impact
106 106  * Requires careful evidence evaluation
107 107  * Multiple valid interpretations possible
108 108  **Publication**: AKEL publishes automatically with standard risk label
109 109  **Audit rate**: Moderate sampling recommended
110 -
111 111  === 3.3 Tier C (Low Risk) ===
112 -
113 113  **Domains**: Definitions, established facts, historical data
114 114  **Characteristics**:
115 -
116 116  * Low potential for harm
117 117  * Well-documented information
118 118  * Clear right/wrong answers typically
119 119  **Publication**: AKEL publishes by default
120 120  **Audit rate**: Lower sampling recommended
121 -
122 122  == 4. Quality Gates ==
123 -
124 124  AKEL applies quality gates before publication. If any fail, claim is **flagged** (not blocked - still published).
125 125  **Quality gates**:
126 -
127 127  * Sufficient evidence extracted (≥2 sources)
128 128  * Sources meet minimum credibility threshold
129 129  * Confidence score calculable
... ... @@ -130,10 +130,8 @@
130 130  * No detected manipulation patterns
131 131  * Claim parseable into testable form
132 132  **Failed gates**: Claim published with flag for moderator review
133 -
134 134  == 5. Automation Levels ==
135 -
136 -{{include reference="Test.FactHarbor pre12 V0\.9\.70.Specification.Diagrams.Automation Level.WebHome"/}}
109 +{{include reference="Test.FactHarbor.Specification.Diagrams.Automation Level.WebHome"/}}
137 137  FactHarbor progresses through automation maturity levels:
138 138  **Release 0.5** (Proof-of-Concept): Tier C only, human review required
139 139  **Release 1.0** (Initial): Tier B/C auto-published, Tier A flagged for review
... ... @@ -152,9 +152,7 @@
152 152  **Action**: May temporarily hide content, ban users, or propose algorithm improvements
153 153  **Does NOT**: Routinely review claims or override verdicts
154 154  See [[Organisational Model>>Test.FactHarbor.Organisation.Organisational-Model]] for moderator role details.
155 -
156 156  == 8. Continuous Improvement ==
157 -
158 158  **Performance monitoring**: Track AKEL accuracy, speed, coverage
159 159  **Issue identification**: Find systematic errors from metrics
160 160  **Algorithm updates**: Deploy improvements to fix patterns
... ... @@ -161,21 +161,15 @@
161 161  **A/B testing**: Validate changes before full rollout
162 162  **Retrospectives**: Learn from failures systematically
163 163  See [[Continuous Improvement>>Test.FactHarbor.Organisation.How-We-Work-Together.Continuous-Improvement]] for improvement cycle.
164 -
165 165  == 9. Scalability ==
166 -
167 167  Automation enables FactHarbor to scale:
168 -
169 169  * **Millions of claims** processable
170 170  * **Consistent quality** at any volume
171 171  * **Cost efficiency** through automation
172 172  * **Rapid iteration** on algorithms
173 173  Without automation: Human review doesn't scale, creates bottlenecks, introduces inconsistency.
174 -
175 175  == 10. Transparency ==
176 -
177 177  All automation is transparent:
178 -
179 179  * **Algorithm parameters** documented
180 180  * **Evaluation criteria** public
181 181  * **Source scoring rules** explicit