Skip to Content
Last modified by Robert Schaub on 2025/12/24 20:35
From version 1.1
edited by Robert Schaub
on 2025/12/21 11:25
Change comment: Imported from XAR
To version 1.8
edited by Robert Schaub
on 2025/12/24 20:35
Change comment: Renamed back-links.

Summary

Details

Page properties
Parent
... ... @@ -1,1 +1,1 @@
1 -Test.FactHarbor.Roadmap.WebHome
1 +Archive.FactHarbor delta for V0\.9\.70.Roadmap.WebHome
Content
... ... @@ -4,7 +4,7 @@
4 4  
5 5  **Success Metric:** <10% hallucination rate, quality gates prevent low-confidence publications
6 6  
7 ----
7 +----
8 8  
9 9  == 1. Overview ==
10 10  
... ... @@ -13,12 +13,13 @@
13 13  **Key Innovation:** Quality validation BEFORE publication, not after
14 14  
15 15  **What We're Proving:**
16 +
16 16  * AKEL can reliably extract factual claims from articles
17 17  * AKEL can generate credible verdicts with proper evidence
18 18  * Quality gates prevent hallucinations and low-confidence outputs
19 19  * Fully automated approach is viable
20 20  
21 ----
22 +----
22 22  
23 23  == 2. Scope ==
24 24  
... ... @@ -42,7 +42,7 @@
42 42  * A/B testing
43 43  * Gates 2 & 3 (Evidence relevance, Scenario coherence)
44 44  
45 ----
46 +----
46 46  
47 47  == 3. Requirements ==
48 48  
... ... @@ -60,6 +60,7 @@
60 60  **Purpose:** Ensure extracted claims are factual assertions, not opinions or predictions
61 61  
62 62  **Validation Checks:**
64 +
63 63  1. **Factual Statement Test:** Can this be verified with evidence?
64 64  2. **Opinion Detection:** Contains hedging language? ("I think", "probably", "best", "worst")
65 65  3. **Specificity Score:** Contains concrete details? (names, numbers, dates, locations)
... ... @@ -66,14 +66,13 @@
66 66  4. **Future Prediction Test:** Makes claims about future events?
67 67  
68 68  **Pass Criteria:**
69 -{{code}}
70 -- isFactual: true
71 +{{code}}- isFactual: true
71 71  - opinionScore: ≤ 0.3
72 72  - specificityScore: ≥ 0.3
73 -- claimType: FACTUAL
74 -{{/code}}
74 +- claimType: FACTUAL{{/code}}
75 75  
76 76  **Action if Failed:**
77 +
77 77  * Flag as "Non-verifiable: Opinion/Prediction/Ambiguous"
78 78  * Do NOT generate scenarios or verdicts
79 79  * Display explanation to user
... ... @@ -80,7 +80,7 @@
80 80  
81 81  **Target:** 0% opinion statements processed as facts
82 82  
83 ----
84 +----
84 84  
85 85  ==== Gate 4: Verdict Confidence Assessment ====
86 86  
... ... @@ -87,6 +87,7 @@
87 87  **Purpose:** Only publish verdicts with sufficient evidence and confidence
88 88  
89 89  **Validation Checks:**
91 +
90 90  1. **Evidence Count:** Minimum 2 independent sources
91 91  2. **Source Quality:** Average reliability ≥ 0.6 (on 0-1 scale)
92 92  3. **Evidence Agreement:** % supporting vs. contradicting ≥ 0.6
... ... @@ -93,8 +93,7 @@
93 93  4. **Uncertainty Factors:** Count of hedging statements in reasoning
94 94  
95 95  **Confidence Tiers:**
96 -{{code}}
97 -HIGH (80-100%):
98 +{{code}}HIGH (80-100%):
98 98   - ≥3 sources
99 99   - ≥0.7 average quality
100 100   - ≥80% agreement
... ... @@ -108,16 +108,16 @@
108 108   - ≥2 sources BUT low quality/agreement
109 109  
110 110  INSUFFICIENT:
111 - - <2 sources → DO NOT PUBLISH
112 -{{/code}}
112 + - <2 sources → DO NOT PUBLISH{{/code}}
113 113  
114 114  **POC1 Publication Rule:**
115 +
115 115  * Minimum **MEDIUM** confidence required
116 116  * Blocked verdicts show "Insufficient Evidence" message
117 117  
118 118  **Target:** 0% verdicts published with <2 sources
119 119  
120 ----
121 +----
121 121  
122 122  === 3.2 Modified FR7: Automated Verdicts (Enhanced) ===
123 123  
... ... @@ -145,6 +145,7 @@
145 145  {{/code}}
146 146  
147 147  **Updated Verdict States:**
149 +
148 148  * PUBLISHED - Passed all gates
149 149  * INSUFFICIENT_EVIDENCE - Failed Gate 4
150 150  * NON_FACTUAL_CLAIM - Failed Gate 1
... ... @@ -151,7 +151,7 @@
151 151  * PROCESSING - In progress
152 152  * ERROR - System failure
153 153  
154 ----
156 +----
155 155  
156 156  === 3.3 Modified FR4: Analysis Summary (Enhanced) ===
157 157  
... ... @@ -177,7 +177,7 @@
177 177   Quality Score: 8.5/10
178 178  {{/code}}
179 179  
180 ----
182 +----
181 181  
182 182  == 4. Success Criteria ==
183 183  
... ... @@ -184,6 +184,7 @@
184 184  POC1 is considered **SUCCESSFUL** if:
185 185  
186 186  **✅ Functional:**
189 +
187 187  * Processes diverse test articles without crashes
188 188  * Generates verdicts for all factual claims
189 189  * Blocks all non-factual claims (0% pass through)
... ... @@ -190,6 +190,7 @@
190 190  * Blocks all insufficient-evidence verdicts (0% with <2 sources)
191 191  
192 192  **✅ Quality:**
196 +
193 193  * Hallucination rate <10% (manual verification)
194 194  * 0 verdicts with <2 sources published
195 195  * 0 opinion statements published as facts
... ... @@ -196,17 +196,19 @@
196 196  * Average quality score ≥7.0/10
197 197  
198 198  **✅ Performance:**
203 +
199 199  * Processing time reasonable for POC demonstration
200 200  * Quality gates execute efficiently
201 201  * UI displays results clearly
202 202  
203 203  **✅ Learnings:**
209 +
204 204  * Identified prompt engineering improvements
205 205  * Documented AKEL strengths/weaknesses
206 206  * Validated threshold values
207 207  * Clear path to POC2 defined
208 208  
209 ----
215 +----
210 210  
211 211  == 5. Decision Gates ==
212 212  
... ... @@ -219,7 +219,7 @@
219 219  
220 220  **Only proceed to POC2 if all success criteria met**
221 221  
222 ----
228 +----
223 223  
224 224  == 6. Architecture Notes ==
225 225  
... ... @@ -239,6 +239,7 @@
239 239  {{/code}}
240 240  
241 241  **POC1 Acceptable Simplifications:**
248 +
242 242  * Single AKEL call (not multi-component pipeline)
243 243  * No scenarios (implicit in verdicts)
244 244  * Basic evidence linking
... ... @@ -245,20 +245,19 @@
245 245  * 2 gates instead of 4
246 246  * No review queue
247 247  
248 -**See:** [[Architecture>>Test.FactHarbor.Specification.Architecture.WebHome]] for details
255 +**See:** [[Architecture>>FactHarbor.Archive.FactHarbor delta for V0\.9\.70.Specification.Architecture.WebHome]] for details
249 249  
250 ----
257 +----
251 251  
252 252  == Related Pages ==
253 253  
254 -* [[Roadmap Overview>>Test.FactHarbor.Roadmap.WebHome]] - All phases
255 -* [[POC2 Requirements>>Test.FactHarbor.Roadmap.POC2.WebHome]] - Next phase
256 -* [[Requirements>>Test.FactHarbor.Specification.Requirements.WebHome]] - Full system requirements
257 -* [[Architecture>>Test.FactHarbor.Specification.Architecture.WebHome]] - System architecture
261 +* [[Roadmap Overview>>FactHarbor.Archive.FactHarbor delta for V0\.9\.70.Roadmap.WebHome]] - All phases
262 +* [[POC2 Requirements>>Archive.FactHarbor delta for V0\.9\.70.Roadmap.POC2.WebHome]] - Next phase
263 +* [[Requirements>>FactHarbor.Archive.FactHarbor delta for V0\.9\.70.Specification.Requirements.WebHome]] - Full system requirements
264 +* [[Architecture>>FactHarbor.Archive.FactHarbor delta for V0\.9\.70.Specification.Architecture.WebHome]] - System architecture
258 258  * [[NFR11 Full Specification>>Test.FactHarbor.Specification.Requirements.WebHome#NFR11]] - Complete quality framework
259 259  
260 ----
267 +----
261 261  
262 262  **Document Status:** ✅ POC1 Specification Complete - Ready for Implementation
263 263  **Version:** V0.9.70
264 -