knowledge graph construction with rebrickable data

lego/.gitignore

@@ -0,0 +1 @@
+*.ttl

lego/lego_graph.ipynb

@@ -0,0 +1,308 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "id": "747b245f",
+   "metadata": {},
+   "source": [
+    "Build the Lego Knwoledge Graph using the sources in `/data`."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 40,
+   "id": "90209948",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from rdflib import Graph, Namespace, XSD, OWL, RDF, RDFS, SKOS, URIRef, Literal\n",
+    "import pandas as pd\n",
+    "from datetime import datetime"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "fe91fa67",
+   "metadata": {},
+   "source": [
+    "Setup the requirements for building a knowledge graph"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 41,
+   "id": "8e573135",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "g = Graph()\n",
+    "thm = Namespace(\"https://th-mannheim.de/\")\n",
+    "THM = Namespace(\"https://th-mannheim.de/ont/\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "d56199d5",
+   "metadata": {},
+   "source": [
+    "# Rebrickable"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "d1e1abb0",
+   "metadata": {},
+   "source": [
+    "![Rebrickable](\\data\\rebrickable\\downloads_schema_v3.png)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 42,
+   "id": "d8a1fe84",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "re_colors = pd.read_csv(\"data/rebrickable/colors.csv\")\n",
+    "re_elements = pd.read_csv(\"data/rebrickable/elements.csv\")\n",
+    "re_inventories = pd.read_csv(\"data/rebrickable/inventories.csv\")\n",
+    "re_inventory_minifigs = pd.read_csv(\"data/rebrickable/inventory_minifigs.csv\")\n",
+    "re_inventory_parts = pd.read_csv(\"data/rebrickable/inventory_parts.csv\")\n",
+    "re_inventory_sets = pd.read_csv(\"data/rebrickable/inventory_sets.csv\")\n",
+    "re_minifigs = pd.read_csv(\"data/rebrickable/minifigs.csv\")\n",
+    "re_part_categories = pd.read_csv(\"data/rebrickable/part_categories.csv\")\n",
+    "re_part_relationships = pd.read_csv(\"data/rebrickable/part_relationships.csv\")\n",
+    "re_parts = pd.read_csv(\"data/rebrickable/parts.csv\")\n",
+    "re_sets = pd.read_csv(\"data/rebrickable/sets.csv\")\n",
+    "re_themes = pd.read_csv(\"data/rebrickable/themes.csv\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "f3677416",
+   "metadata": {},
+   "source": [
+    "Colors"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 43,
+   "id": "ae505704",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "for color in re_colors.itertuples(index=False):\n",
+    "    color_ref = thm[f\"colors/{color.id}\"]\n",
+    "\n",
+    "    g.add((color_ref, RDFS.label, Literal(color.name, lang=\"en\")))\n",
+    "    g.add((color_ref, THM.color, Literal(color.rgb)))\n",
+    "    g.add((color_ref, THM.is_transparent, Literal(color.is_trans, datatype=XSD.boolean)))\n",
+    "    \n",
+    "    if not pd.isna(color.y1):\n",
+    "        # First appearance\n",
+    "        g.add((color_ref, THM.first_year, Literal(datetime(year = int(color.y1), month=1, day=1))))\n",
+    "    if not pd.isna(color.y2):\n",
+    "        # Last appearance\n",
+    "        g.add((color_ref, THM.last_year, Literal(datetime(year = int(color.y2), month=1, day=1))))\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "e27b2bc4",
+   "metadata": {},
+   "source": [
+    "Part Categories"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 44,
+   "id": "fb9e17d6",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "for part_category in re_part_categories.itertuples(index=False):\n",
+    "    part_category_ref = thm[f\"part_category/{part_category.id}\"]\n",
+    "\n",
+    "    g.add((part_category_ref, RDFS.label, Literal(part_category_ref, lang=\"en\")))"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "ea32849b",
+   "metadata": {},
+   "source": [
+    "Parts"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 45,
+   "id": "8fdb080e",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "for part in re_parts.itertuples(index=False):\n",
+    "    part_ref = thm[f\"part/{part.part_num}\"]\n",
+    "\n",
+    "    g.add((part_ref, RDFS.label, Literal(part.name, lang=\"en\")))\n",
+    "    g.add((part_ref, THM.part_category, thm[f\"part_category/{part.part_cat_id}\"]))\n",
+    "    g.add((part_ref, THM.part_material, Literal(part.part_material)))"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "fcaadd84",
+   "metadata": {},
+   "source": [
+    "Elements"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 46,
+   "id": "579b1d67",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "for element in re_elements.itertuples(index=False):\n",
+    "    part_ref = thm[f\"part/{element.part_num}\"]\n",
+    "    color_ref = thm[f\"colors/{element.color_id}\"]\n",
+    "\n",
+    "    g.add((part_ref, THM.has_color, color_ref))"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "44dae336",
+   "metadata": {},
+   "source": [
+    "Part Relationships"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 47,
+   "id": "00db079a",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "for part_relationship in re_part_relationships.itertuples(index=False):\n",
+    "    part_ref_parent = thm[f\"part/{part_relationship.parent_part_num}\"]\n",
+    "    part_ref_child = thm[f\"part/{part_relationship.child_part_num}\"]\n",
+    "\n",
+    "    g.add((part_ref_parent, THM.has_child, part_ref_child))"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "19dc64b8",
+   "metadata": {},
+   "source": [
+    "Themes"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "1a529aae",
+   "metadata": {},
+   "outputs": [
+    {
+     "ename": "SyntaxError",
+     "evalue": "f-string: unmatched ')' (1024367582.py, line 2)",
+     "output_type": "error",
+     "traceback": [
+      "  \u001b[36mCell\u001b[39m\u001b[36m \u001b[39m\u001b[32mIn[48]\u001b[39m\u001b[32m, line 2\u001b[39m\n\u001b[31m    \u001b[39m\u001b[31mtheme_ref = thm[f\"theme/{int(theme.id))}\"]\u001b[39m\n                                          ^\n\u001b[31mSyntaxError\u001b[39m\u001b[31m:\u001b[39m f-string: unmatched ')'\n"
+     ]
+    }
+   ],
+   "source": [
+    "for theme in re_themes.itertuples(index=False):\n",
+    "    theme_ref = thm[f\"theme/{int(theme.id)}\"]\n",
+    "\n",
+    "    g.add((theme_ref, RDFS.label, Literal(theme.name)))\n",
+    "\n",
+    "    if not pd.isna(theme.parent_id):\n",
+    "        g.add((theme_ref, THM.parent_theme, thm[f\"theme/{int(theme.parent_id)}\"]))"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "bfab0c73",
+   "metadata": {},
+   "source": [
+    "Serialize the graph in turtle format"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "2abd6894",
+   "metadata": {},
+   "source": [
+    "```\n",
+    "                             ___-------___\n",
+    "                         _-~~             ~~-_\n",
+    "                      _-~                    /~-_\n",
+    "   /^\\__/^\\         /~  \\                   /    \\\n",
+    " /|  O|| O|        /      \\_______________/        \\\n",
+    "| |___||__|      /       /                \\          \\\n",
+    "|          \\    /      /                    \\          \\\n",
+    "|   (_______) /______/                        \\_________ \\\n",
+    "|         / /         \\                      /            \\\n",
+    " \\         \\^\\\\         \\                  /               \\     /\n",
+    "   \\         ||           \\______________/      _-_       //\\__//\n",
+    "     \\       ||------_-~~-_ ------------- \\ --/~   ~\\    || __/\n",
+    "       ~-----||====/~     |==================|       |/~~~~~\n",
+    "        (_(__/  ./     /                    \\_\\      \\.\n",
+    "               (_(___/                         \\_____)_)\n",
+    "```"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "1a30bff8",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<Graph identifier=Nf661b2e682c043188ddd822a6bca246c (<class 'rdflib.graph.Graph'>)>"
+      ]
+     },
+     "execution_count": 36,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "g.bind(\"thmont\", THM)\n",
+    "\n",
+    "g.serialize(\"lego_graph.ttl\", format=\"turtle\")"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "venv (3.14.4)",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.14.4"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}

lego/paper/KGR_paper1_lego.tex

@@ -142,7 +142,7 @@
 		\subsection{Merlins Steine}
 		
 		\textit{Merlins Steine} ist eine Website, die sich auf die Berichterstattung von Reviews zu Klemmbausteinen spezifiziert. Um auf entsprechende Sets zu referenzieren beinhaltet \textit{Merlins Steine} eine Datenbank an Klemmbausteinen, die über 30000 Sets unterschiedlicher Marken enthält. Merlins Steine finanziert sich über Affiliate Marketing. \cite{SetDatenbankSetDB}.\\
-		\textit{Merlins Steine} wurde ausgewählt, da Sets von anderen Marken im Vergleich zu vorherig genannten Quellen aufgelistet werden.
+		\textit{Merlins Steine} wurde ausgewählt, da Sets von anderen Marken im Vergleich zu vorherig genannten Quellen aufgelistet werden. Der Datensatz enthält Informationen zum \ac{UVP}, Teileanzahl, Release-Jahr, Thema Modelldimensionen und des Herstellers.
 		
 		\begin{table}[H]
 			\begin{tabularx}{\columnwidth}{@{}ll@{}}