Remove table and image data from manual annotations, keep selectable text

Ground-truth samples now contain only selectable text and captions:
pipe/inline table rows and figure-data grids are dropped from _raw.md
and each sub-json section, while table/figure captions, prose, headings,
bullets and contact/venue lines are kept. Affects 10890106, 16083265,
4737132, 8228476, 8228568.

Author: jimnoneill

manual_poster_annotation/10890106/10890106_raw.md

@@ -55,12 +55,6 @@ Figure 3. A) hiPSCs and MUTZ-3 cells were differentiated into IECs and mDCs sepa
 
 Table 1. Physicochemical data of the micro/nanoplastics for the cell exposure experiment.
 
-| Abbreviation | Main polymer composition | Additives | Particle size distribution | Particle shape |
-|--------------|-------------------------|-----------|----------------------------|----------------|
-| HDPE269 | High density polyethylene | No | X₁₀;₃ 1.8 µm / X₅₀;₃ 4.9 µm / X₉₀;₃ 8.7 µm | Round |
-| PET_c000 | Polyethylene terephthalate | No | X₁₀;₃ 57 nm / X₅₀;₃ 90 nm / X₉₀;₃ 144 nm | Round |
-| PS394Eu | Europium doped polystyrene | No | 300 nm | Round |
-
 Figure 4. Gene expression in iDCs after 48h exposure to micro/nano plastics at three different concentrations. A) Effects of the micro/nanoplastics on gene expression of markers for maturation of iDCs into mDCs and B) on cytokine expression. No significant effects were observed after 24h exposure to the micro/nanoplastics and no cytotoxicity was observed (data not shown). After 48h exposure to the PS394Eu plastics mild effects on cytokine expression were observed.
 
 ## Conclusions

manual_poster_annotation/16083265/16083265_raw.md

@@ -48,13 +48,6 @@ Figure 1: Representative scheme of a) outer and b) inner side of microfluidic ch
 
 Table 1: Drug loading into NPs and micelles.
 
-| Type of Syringes | Final Volume (mL) | Total Flow Rate (TFR) (mL/min) | Flow Rate Ratio (FRR) (aqueous/organic phase) |
-|------------------|-------------------|-------------------------------|-----------------------------------------------|
-| PLGA NPs | Terumo 5 mL | 2 | 8 | 4:1 |
-| | Terumo 5 mL | 2 | 8 | 4:1 |
-| PLA/PEG NPs | Terumo 5 mL | 2 | 3 | 3:1 |
-| | Terumo 5 mL | 2 | 3 | 3:1 |
-
 ## RESULTS
 
 1. Physical Properties of NPs
@@ -69,13 +62,6 @@ Figure 3: TEM images of a) PLGA NPs, b) CURC loaded PLGA NPs, c) PLA-PEG micelle
 
 Table 2: Drug loading into NPs and micelles.
 
-| Loading | PLGA | PLA/PEG |
-|---------|------|---------|
-| EE % | | |
-| Curcumin | 61.91 ± 1.86 | 13.74 ± 0.97 |
-| Payload (% µg/mg) | | |
-| | 2.81 ± 0.81 | 2.51 ± 0.57 |
-
 Higher Encapsulation Efficiency (EE%)
 
 Figure 4: Release profiles of CURC from PLGA NPs and PLA/PEG micelles.

manual_poster_annotation/16083265/16083265_sub-json.json

@@ -85,11 +85,11 @@
       },
       {
         "sectionTitle": "METHODS",
-        "sectionContent": "• An ad hoc custom-built device was used, comprising a pumping module with two syringe pumps and an infusion setup designed to place a Passive Herringbone Mixer (PHBM) chip.\n\nTable 1: Drug loading into NPs and micelles.\n\nType of Syringes | Final Volume (mL) | Total Flow Rate (TFR) (mL/min) | Flow Rate Ratio (FRR) (aqueous/organic phase)\nPLGA NPs | Terumo 5 mL | 2 | 8 | 4:1\n         | Terumo 5 mL | 2 | 8 | 4:1\nPLA/PEG NPs | Terumo 5 mL | 2 | 3 | 3:1\n            | Terumo 5 mL | 2 | 3 | 3:1"
+        "sectionContent": "• An ad hoc custom-built device was used, comprising a pumping module with two syringe pumps and an infusion setup designed to place a Passive Herringbone Mixer (PHBM) chip.\n\nTable 1: Drug loading into NPs and micelles.\n"
       },
       {
         "sectionTitle": "RESULTS",
-        "sectionContent": "1. Physical Properties of NPs\n\n• NPs had uniform size distribution with PDI ≤ 0.2, consistent with TEM observations.\n\nTable 2: Drug loading into NPs and micelles.\n\nEE %\nLoading | PLGA | PLA/PEG\nCurcumin | 61.91 ± 1.86 | 13.74 ± 0.97\n\nPayload (% µg/mg)\n2.81 ± 0.81 | 2.51 ± 0.57\n\nHigher Encapsulation Efficiency (EE%)\n\nSlower Release Rate\n\n2. Release Kinetics of NPs\n\n3. Cytotoxicity & Antimicrobial Activity"
+        "sectionContent": "1. Physical Properties of NPs\n\n• NPs had uniform size distribution with PDI ≤ 0.2, consistent with TEM observations.\n\nTable 2: Drug loading into NPs and micelles.\n\nEE %\n\nPayload (% µg/mg)\n\nHigher Encapsulation Efficiency (EE%)\n\nSlower Release Rate\n\n2. Release Kinetics of NPs\n\n3. Cytotoxicity & Antimicrobial Activity"
       },
       {
         "sectionTitle": "CONCLUSION",
@@ -152,4 +152,4 @@
   "formats": [
     "application/pdf"
   ]
-}
+}

manual_poster_annotation/4737132/4737132_raw.md

@@ -40,20 +40,6 @@ Was ist wichtig? – Schlüsselstellen in der Literatur
 
 Zitierte Stellen aus Interpretationstexten
 
-w1 w2 w3 w4 w5 w6 w7 w8 w9 w10
-
-w3 w4 w5 w6 w7 w8 w9
-
-w2 w3 w4
-
-w1 w2 w3
-
-w8 w9 w10
-
-Literatur w1 | w2 | w3 | w4 | w5 w6 w7 | w8 w9 | w10
-
-Zitierhäufigkeit 2 | 3 | 4 | 3 | 2 | 3 | 2
-
 ## Comic
 
 https://xkcd.com/1262 (Stand: 12.03.2021)
@@ -62,28 +48,8 @@ https://xkcd.com/1262 (Stand: 12.03.2021)
 
 ### 1. Positionen in Text 1 erfassen
 
-| Wortfolge | Startpositionen |
-|-----------|-----------------|
-| w1 w2 w3 | [0] |
-| w2 w3 w4 | [1] |
-| ... ... ... | ... |
-| w8 w9 w10 | [7] |
-| w9 w10 w11 | [8] |
-| w10 w11 w12 | [9, 35] |
-| w33 w34 w35 | [32] |
-
 ### 2. Forward references erstellen
 
-| Text 1 Startpos. | Text 2 Startpos. |
-|------------------|------------------|
-| 0 | [] |
-| 1 | [] |
-| 7 | [75, 120] |
-| 8 | [121] |
-| 9 | [77, 122] |
-| 35 | [77, 122] |
-| 32 | [145] |
-
 ### 3. Gefundene Matches erweitern
 
 • Ellipsen "überspringen"

manual_poster_annotation/4737132/4737132_sub-json.json

@@ -49,7 +49,7 @@
       },
       {
         "sectionTitle": "Segmentierung",
-        "sectionContent": "Zitierte Stellen aus Interpretationstexten\n\nw1 w2 w3 w4 w5 w6 w7 w8 w9 w10\nw3 w4 w5 w6 w7 w8 w9\nw2 w3 w4\nw2 w3\nw8 w9 w10\n\nLiteratur w1 | w2 | w3 | w4 | w5 w6 w7 | w8 w9 | w10\nZitierhäufigkeit 2 | 3 | 4 | 3 | 2 | 3 | 2"
+        "sectionContent": "Zitierte Stellen aus Interpretationstexten"
       },
       {
         "sectionTitle": "Comic",
@@ -86,4 +86,4 @@
   "formats": [
     "image/jpeg"
   ]
-}
+}

manual_poster_annotation/8228476/8228476_raw.md

@@ -19,12 +19,6 @@ A medieval Hebrew poem from the Worms Prayerbook, Germany, 13th Century, The Nat
 • 180 labeled texts, 7,000 sentences, 45,000 words.
 • 15% of the words are metaphors.
 
-| Hebrew | English | Meaning |
-|--------|---------|---------|
-| טבענו בגזרות, כשטופי ים | We drowned in decrees, like sea washed | There are too many decrees |
-| הצית נשמה ויבער נר | Ignite a soul, fire a candle | Create a soul |
-| עד דור עשירי לא עשו פרי | Until the tenth generation they did not bear fruit | For ten generations, they did no good deeds |
-
 ### Inter Annotator Agreement
 
 Inter-annotator agreement evaluates how consistently different annotators label data for NLP research (measured in Cohen's kappa). The score for our dataset is 0.618 (considered substantial).
@@ -61,12 +55,6 @@ Tokenizer
 
 Trained on additional prose and poetic medieval texts (Project Ben-Yehuda*) + Piyyut dataset
 
-| Model | Data Size (words) | Data Type |
-|-------|-------------------|-----------|
-| BEREL | 220 Million | Medieval |
-| AlephBERT | 2 Billion | Modern |
-| AlephBERT Adapted | 2 Billion + 10 Million | Modern + Medieval |
-
 *Project Ben-Yehuda is a volunteer-based online library of free Hebrew literature.
 
 ## Results

manual_poster_annotation/8228568/8228568_raw.md

@@ -42,20 +42,8 @@ Researchers' self-confidence in technical literacy and the nature of research qu
 
 ### Table 1. Why do researchers self-identify as digital humanists?
 
-| Reason | Participant (n=8) | Example |
-|--------|-------------------|---------|
-| Competent in digital methods (programming and coding in particular) | P2, P6, P7, P9, P11 | "I have a background in anthropology and museum studies, and I know coding and know how to do data analysis. …" (P9) "I can code and I use computational and quantitative methods to study literature." (P11) |
-| Active working, teaching, or supervising students in DH | P4, P10, P14 | "I self-identify as a digital humanist in the same way as I self-identify as an information designer, a programmer, and a statistician. I do research in digital humanities, so I guess technically I can say I am a digital humanist." (P4) |
-
 ### Table 2. Why do researchers NOT self-identify as digital humanists?
 
-| Reason | Participant (n=6) | Example |
-|--------|-------------------|---------|
-| Not a "humanist" | P12, P13 | "I wouldn't think I'm a humanist because my research question is not a humanities research question. I approach cultural heritage research from information science and design perspectives…. although I think what I'm doing is digital humanities work." (P13) |
-| Concerned with technical skills (programming in particular) | P8 | "I just started learning programming and digital methods. Being a digital humanist will be my final goal, but I am not there yet." (P8) |
-| DH does not raise new research questions | P3 | "…new constructs such as "DH" did not generate fundamentally new questions distinct from those in conventional or analog humanities." (P3) |
-| Digital methods are not necessary | P1, P5 | "My research questions are not ultimately bound with digital methods; they can be addressed with digital methods, but not necessarily have to be." (P5) |
-
 ## References
 
 Alvarado, R. C. (2012). The digital humanities situation. In M. Gold (Ed.), Debates in the Digital Humanities. University of Minnesota Press.