scimapper 1.0.1__py3-none-any.whl

core/visualizer.py

@@ -0,0 +1,271 @@
+"""
+AutoSMS Framework - Módulo de Visualización Avanzada (Visualizer)
+===============================================================
+Construye visualizaciones interactivas de última generación en Plotly:
+1. Vista Dual: Mapa Conceptual 2D de Dispersión + Capa de Contorno de Densidad (KDE).
+2. Panel Lateral de Evolución Temporal (2018–2026) con cálculo de tasas de crecimiento.
+3. Selector interactivo de modos de visualización (Híbrido, Puntos, Densidad Topográfica).
+"""
+
+import os
+import pandas as pd
+import numpy as np
+import plotly.graph_objects as go
+from plotly.subplots import make_subplots
+
+class AutoVisualizer:
+    def __init__(self, config: dict):
+        self.config = config
+        self.output_dir = config.get("project_metadata", {}).get("output_dir", "./output")
+        self.title = config.get("project_metadata", {}).get("title", "Mapeo Sistemático de Literatura")
+
+    def build_interactive_map(self, df: pd.DataFrame, topics_info: dict) -> str:
+        print("\n🌐 Construyendo Dashboard Dual Interactivo (Mapa 2D KDE + Trayectorias Temporales)...")
+
+        colores = [
+            "#1f77b4", "#ff7f0e", "#2ca02c", "#d62728",
+            "#9467bd", "#8c564b", "#e377c2", "#7f7f7f", "#bcbd22", "#17becf"
+        ]
+
+        # Crear figura con 2 subplots en paralelo: Mapa 2D (58%) y Líneas Temporales (42%)
+        fig = make_subplots(
+            rows=1, cols=2,
+            column_widths=[0.58, 0.42],
+            subplot_titles=[
+                "<b>Mapa Semántico 2D y Densidad de Investigación (t-SNE + KDE)</b>",
+                "<b>Dinámica y Trayectoria Temporal Anual</b>"
+            ],
+            horizontal_spacing=0.08
+        )
+
+        # 1. Capa Base: Mapa de Contorno / Densidad (KDE 2D)
+        kde_trace = go.Histogram2dContour(
+            x=df["coord_x"],
+            y=df["coord_y"],
+            name="Densidad Topográfica (KDE)",
+            colorscale="Blues",
+            reversescale=False,
+            showscale=False,
+            ncontours=22,
+            opacity=0.30,
+            hoverinfo="skip"
+        )
+        fig.add_trace(kde_trace, row=1, col=1)
+
+        # 2. Capas de Clusters (Envolventes Convexas, Puntos y Centroides)
+        from scipy.spatial import ConvexHull
+
+        def hex_to_rgb(hex_str):
+            hex_str = hex_str.lstrip('#')
+            return tuple(int(hex_str[i:i+2], 16) for i in (0, 2, 4))
+
+        unique_topics = sorted(df["topic_id"].unique())
+        total_docs = len(df)
+        hull_trace_indices = []
+        scatter_trace_indices = []
+        centroid_trace_indices = []
+
+        for idx, tid in enumerate(unique_topics):
+            sub = df[df["topic_id"] == tid]
+            color = colores[idx % len(colores)]
+            r, g, b = hex_to_rgb(color)
+            t_info = topics_info.get(str(tid), topics_info.get(tid, {}))
+            t_name = t_info.get("assigned_name", f"Tópico {tid}")
+            
+            # Dinámica temporal para tooltip
+            t_dyn = t_info.get("temporal_dynamics", {})
+            badge = t_dyn.get("badge", "⚖️ Maduro")
+            cagr = t_dyn.get("cagr_percent", 0.0)
+            kws = ", ".join(t_info.get("keywords", [])[:5])
+
+            pts = sub[["coord_x", "coord_y"]].values
+            center = np.mean(pts, axis=0)
+            cx, cy = float(center[0]), float(center[1])
+
+            # A. Envolvente Convexa del Cluster
+            dists = np.linalg.norm(pts - center, axis=1)
+            mask_core = dists <= np.percentile(dists, 92)
+            pts_core = pts[mask_core]
+
+            if len(pts_core) >= 3:
+                hull = ConvexHull(pts_core)
+                hull_pts = pts_core[hull.vertices]
+                hull_pts = np.vstack([hull_pts, hull_pts[0]])
+                hx = hull_pts[:, 0]
+                hy = hull_pts[:, 1]
+            else:
+                hx = [cx]
+                hy = [cy]
+
+            hull_trace = go.Scatter(
+                x=hx,
+                y=hy,
+                mode="lines",
+                fill="toself",
+                fillcolor=f"rgba({r}, {g}, {b}, 0.16)",
+                line=dict(color=f"rgba({r}, {g}, {b}, 0.85)", width=2, dash="dash", shape="spline", smoothing=1.3),
+                name=f"Área {badge} {t_name}",
+                legendgroup=f"topic_{tid}",
+                showlegend=False,
+                hoverinfo="skip"
+            )
+            fig.add_trace(hull_trace, row=1, col=1)
+            hull_trace_indices.append(len(fig.data) - 1)
+
+            # B. Puntos individuales
+            hover_texts = []
+            for _, r_doc in sub.iterrows():
+                snip = (r_doc['abstract'][:240] + "...") if len(r_doc['abstract']) > 240 else r_doc['abstract']
+                txt = (
+                    f"<b>{r_doc['title']}</b><br>"
+                    f"<i>Autores:</i> {r_doc['authors']}<br>"
+                    f"<i>Año:</i> {r_doc['publication_year']} | <i>Venue:</i> {r_doc['venue']}<br>"
+                    f"<i>Citas:</i> {r_doc.get('cited_by_count', 0)} | <b>Estado:</b> {badge} (CAGR: {cagr}%)<br>"
+                    f"<hr style='margin:4px 0;'>"
+                    f"<b>Abstract:</b> {snip}"
+                )
+                hover_texts.append(txt)
+
+            scatter_trace = go.Scatter(
+                x=sub["coord_x"],
+                y=sub["coord_y"],
+                mode="markers",
+                name=f"{badge} {t_name} ({len(sub)})",
+                text=hover_texts,
+                hoverinfo="text",
+                legendgroup=f"topic_{tid}",
+                marker=dict(size=8, color=color, opacity=0.88, line=dict(width=1, color="#222222"))
+            )
+            fig.add_trace(scatter_trace, row=1, col=1)
+            scatter_trace_indices.append(len(fig.data) - 1)
+
+            # C. Centroide y Etiqueta de Cluster
+            centroid_hover = (
+                f"<b>🏛️ CLUSTER {tid}: {t_name}</b><br>"
+                f"<b>Estado:</b> {badge} (CAGR: {cagr}%)<br>"
+                f"<b>Volumen:</b> {len(sub)} artículos ({round(len(sub)/total_docs*100, 1)}%)<br>"
+                f"<b>Palabras Clave (c-TF-IDF):</b> {kws}<br>"
+                f"<i>Centroide: ({cx:.2f}, {cy:.2f})</i>"
+            )
+
+            centroid_trace = go.Scatter(
+                x=[cx],
+                y=[cy],
+                mode="markers+text",
+                text=[f"<b>C{tid}</b>"],
+                textposition="top center",
+                textfont=dict(size=12, color="#0f172a", family="Arial Black, Segoe UI, sans-serif"),
+                hovertext=[centroid_hover],
+                hoverinfo="text",
+                legendgroup=f"topic_{tid}",
+                showlegend=False,
+                marker=dict(size=16, symbol="diamond", color=color, line=dict(width=2, color="#ffffff"))
+            )
+            fig.add_trace(centroid_trace, row=1, col=1)
+            centroid_trace_indices.append(len(fig.data) - 1)
+
+        # 3. Subplot 2: Evolución Temporal Anual (Líneas de tendencia)
+        years_series = pd.to_numeric(df["publication_year"], errors="coerce").dropna().astype(int)
+        min_year = int(years_series.min()) if not years_series.empty else 2018
+        max_year = int(years_series.max()) if not years_series.empty else 2026
+        all_years = list(range(min_year, max_year + 1))
+
+        for idx, tid in enumerate(unique_topics):
+            color = colores[idx % len(colores)]
+            t_info = topics_info.get(str(tid), topics_info.get(tid, {}))
+            t_name = t_info.get("assigned_name", f"Tópico {tid}")
+            t_dyn = t_info.get("temporal_dynamics", {})
+            y_dist = t_dyn.get("yearly_distribution", {})
+
+            y_values = [y_dist.get(y, y_dist.get(str(y), 0)) for y in all_years]
+
+            line_trace = go.Scatter(
+                x=all_years,
+                y=y_values,
+                mode="lines+markers",
+                name=f"{t_name}",
+                legendgroup=f"topic_{tid}",
+                showlegend=False,
+                line=dict(color=color, width=2.8),
+                marker=dict(size=7, color=color),
+                hovertemplate=f"<b>{t_name}</b><br>Año %{{x}}: %{{y}} artículos<extra></extra>"
+            )
+            fig.add_trace(line_trace, row=1, col=2)
+
+        # 4. Botones de Control para Alternar Modos de Visualización (Updatemenus)
+        total_traces = len(fig.data)
+        
+        vis_all = [True] * total_traces
+        vis_regions = [True] * total_traces
+        for p_idx in scatter_trace_indices:
+            vis_regions[p_idx] = False
+            
+        vis_points = [False] * total_traces
+        for p_idx in scatter_trace_indices + centroid_trace_indices:
+            vis_points[p_idx] = True
+        for l_idx in range(len(fig.data) - len(unique_topics), len(fig.data)):
+            vis_points[l_idx] = True
+
+        fig.update_layout(
+            autosize=True,
+            height=860,
+            template="plotly_white",
+            updatemenus=[
+                dict(
+                    type="buttons",
+                    direction="right",
+                    x=1.0,
+                    xanchor="right",
+                    y=1.14,
+                    showactive=True,
+                    bgcolor="#F8FAFC",
+                    bordercolor="#CBD5E1",
+                    font=dict(size=11, color="#1E293B"),
+                    buttons=[
+                        dict(label="🌟 Vista Completa (Regiones + Puntos + Densidad)", method="update", args=[{"visible": vis_all}]),
+                        dict(label="🛡️ Solo Regiones y Fronteras de Clusters", method="update", args=[{"visible": vis_regions}]),
+                        dict(label="📍 Solo Puntos de Artículos", method="update", args=[{"visible": vis_points}])
+                    ]
+                )
+            ],
+            title={
+                'text': f"<b>{self.title}</b><br><span style='font-size:13px; color:#64748B; font-weight:normal;'>Mapeo Semántico 2D (t-SNE) + Densidad Topográfica (KDE) + Radar de Dinámica Temporal | {len(df)} Artículos Indexados</span>",
+                'x': 0.01,
+                'xanchor': 'left',
+                'y': 0.98,
+                'font': dict(size=17, color="#0F172A", family="Segoe UI, Inter, Arial, sans-serif")
+            },
+            xaxis=dict(title="Dimensión Semántica 1 (t-SNE)", showgrid=True, gridcolor="#F1F5F9", zeroline=False),
+            yaxis=dict(title="Dimensión Semántica 2 (t-SNE)", showgrid=True, gridcolor="#F1F5F9", zeroline=False),
+            xaxis2=dict(
+                title="Año de Publicación",
+                dtick=1,
+                range=[min_year - 0.5, max_year + 0.5],
+                showgrid=True,
+                gridcolor="#F1F5F9"
+            ),
+            yaxis2=dict(title="Volumen de Publicaciones (N)", showgrid=True, gridcolor="#F1F5F9"),
+            legend=dict(
+                title=dict(text="<b>Taxonomías y Estado de Madurez</b>", font=dict(size=12, color="#0F172A")),
+                orientation="h",
+                y=-0.16,
+                x=0.5,
+                xanchor="center",
+                bgcolor="rgba(255,255,255,0.95)",
+                bordercolor="#CBD5E1",
+                borderwidth=1,
+                font=dict(size=11, color="#334155")
+            ),
+            margin=dict(l=60, r=50, t=130, b=120)
+        )
+
+        out_html = os.path.join(self.output_dir, "mapa_interactivo.html")
+        fig.write_html(out_html, config={"responsive": True, "displayModeBar": True})
+        print(f"✔ Dashboard Dual interactivo generado en: {out_html}")
+        return out_html
+
+    def save_final_dataset(self, df: pd.DataFrame) -> str:
+        out_csv = os.path.join(self.output_dir, "corpus_con_topicos.csv")
+        df.to_csv(out_csv, index=False, encoding="utf-8-sig")
+        print(f"✔ Dataset consolidado guardado en: {out_csv}")
+        return out_csv

scimapper-1.0.1.dist-info/METADATA

@@ -0,0 +1,252 @@
+Metadata-Version: 2.4
+Name: scimapper
+Version: 1.0.1
+Summary: An automated scientific mapping and systematic literature review framework with dense semantic clustering and PRISMA 2020 compliance.
+Author-email: Gustavo Andrés Futo <gustavoandresfu@gmail.com>
+License: MIT
+Project-URL: Homepage, https://github.com/gustavoandresfu/SciMapper
+Project-URL: Bug Tracker, https://github.com/gustavoandresfu/SciMapper/issues
+Keywords: systematic-literature-review,systematic-mapping-study,scimapper,scientific-mapping,topic-modeling,sentence-transformers,bertopic,prisma-2020,openalex,evidence-synthesis,nlp
+Classifier: Programming Language :: Python :: 3
+Classifier: License :: OSI Approved :: MIT License
+Classifier: Operating System :: OS Independent
+Classifier: Topic :: Scientific/Engineering :: Information Analysis
+Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
+Classifier: Intended Audience :: Science/Research
+Requires-Python: >=3.9
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Provides-Extra: dev
+Requires-Dist: pytest>=7.0.0; extra == "dev"
+Requires-Dist: ruff>=0.4.0; extra == "dev"
+Provides-Extra: web
+Requires-Dist: streamlit>=1.30.0; extra == "web"
+Provides-Extra: dependencies
+Requires-Dist: requests>=2.28.0; extra == "dependencies"
+Requires-Dist: pandas>=1.5.0; extra == "dependencies"
+Requires-Dist: numpy>=1.23.0; extra == "dependencies"
+Requires-Dist: scikit-learn>=1.2.0; extra == "dependencies"
+Requires-Dist: scipy>=1.10.0; extra == "dependencies"
+Requires-Dist: sentence-transformers>=2.2.0; extra == "dependencies"
+Requires-Dist: torch>=2.0.0; extra == "dependencies"
+Requires-Dist: plotly>=5.13.0; extra == "dependencies"
+Requires-Dist: python-docx>=0.8.11; extra == "dependencies"
+Requires-Dist: tqdm>=4.65.0; extra == "dependencies"
+Dynamic: license-file
+
+<div align="center">
+
+# 🚀 SciMapper
+### Automated Scientific Mapping & Systematic Literature Reviews with Dense Semantic Clustering
+
+[![Python 3.9+](https://img.shields.io/badge/python-3.9+-blue.svg?logo=python&logoColor=white)](https://www.python.org/)
+[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
+[![PRISMA 2020](https://img.shields.io/badge/PRISMA-2020%20Compliant-green.svg)](https://www.prisma-statement.org/)
+[![OpenAlex API](https://img.shields.io/badge/Data%20Source-OpenAlex%20API-orange.svg)](https://openalex.org/)
+[![NLP Engine](https://img.shields.io/badge/NLP-Sentence--Transformers-purple.svg)](https://www.sbert.net/)
+
+**An end-to-end, domain-agnostic framework designed to conduct reproducible Systematic Literature Reviews (SLR) and Systematic Mapping Studies (SMS).**  
+It automates academic corpus harvesting, deduplication, dense semantic embedding, unsupervised clustering, taxonomic discovery, interactive 2D cartography, and manuscript drafting—while keeping **researchers firmly in the loop**.
+
+🌐 **Languages:** **English** • [Español (Guía en Castellano)](README_ES.md)
+
+[Quickstart](#-quickstart-in-3-steps) • [Architecture](#-framework-architecture) • [Comparison](#-comparison-with-the-state-of-the-art) • [Theoretical Foundation](#-theoretical-foundations) • [Citation](#-citation)
+
+</div>
+
+---
+
+## 🌟 Key Features
+
+* 📡 **Automated Scholarly Harvesting:** Direct connection to the **OpenAlex API** (*Polite Pool* with 250M+ works) executing boolean queries and reconstructing inverted-index abstracts in $O(N \log N)$.
+* 🧹 **Bidirectional Deduplication:** Collapses duplicates across arXiv, IEEE, ACM, Springer, and PubMed via canonical lowercase DOI and normalized lexical title hashing.
+* 🧠 **Domain-Specific Transformer Presets:** Plug-and-play dense representations via `Sentence-Transformers`:
+  * `software_engineering`: `sentence-transformers/all-MiniLM-L6-v2`
+  * `biomedical_health`: `microsoft/BiomedNLP-PubMedBERT-base-uncased-abstract`
+  * `scientific_general`: `allenai/scibert_scivocab_uncased`
+  * `finance_business`: `ProsusAI/finbert`
+  * `multilingual`: `sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2`
+* 🔬 **Unsupervised Topic Discovery ($c$-TF-IDF):** Discovers latent research themes without generative hallucination, extracting mathematically distinctive n-grams using *Class-based TF-IDF*.
+* 🗺️ **Interactive Cartography & Cluster Hulls:** Produces publication-grade 2D HTML dashboards (Plotly) featuring:
+  * Regional boundaries via **Convex Hulls** with smoothed spline contours.
+  * Cluster centroids with real-time metadata cards ($N$, percentage, top keywords).
+  * Layer switching (*Full View*, *Cluster Boundaries Only*, *Paper Scatter Points*).
+* 📈 **Temporal Dynamics & Hot Topic Radar:** Computes Compound Annual Growth Rates (CAGR) per cluster to classify topics into 🚀 *Emerging*, ⚖️ *Mature*, or 📉 *Declining*.
+* 🛡️ **Mathematical PRISMA 2020 Audit:** Objective isolation of out-of-scope or foreign language clusters as documented exclusion criteria.
+* 📝 **Automated Manuscript Generation:** Compiles an executive Microsoft Word (`.docx`) template complete with cover page, research questions (RQs), PRISMA summary tables, taxonomy distribution, and structured discussion sections.
+
+---
+
+## 🏗️ Framework Architecture
+
+```mermaid
+graph TD
+    A["⚙️ 1. Project Configuration<br><code>scimapper_config.json</code>"] --> B["📡 2. Harvester & Deduplicator<br><code>collector.py (OpenAlex API)</code>"]
+    B --> C["🧠 3. Semantic Embeddings<br><code>embedder.py (SBERT / MiniLM)</code>"]
+    C --> D["🔬 4. Clustering & c-TF-IDF<br><code>clusterer.py (PCA + t-SNE)</code>"]
+    D --> E1["🗺️ 5a. Interactive Cartography<br><code>visualizer.py (Plotly 2D + Hulls)</code>"]
+    D --> E2["⚖️ 5b. Human-in-the-Loop Review<br><code>discovered_topics.json</code>"]
+    E2 --> F["📝 6. Automated Manuscript<br><code>doc_builder.py (.docx Output)</code>"]
+```
+
+---
+
+## ⚡ Quickstart in 3 Steps
+
+### 1. Installation
+
+Clone the repository and install the dependencies:
+
+```bash
+git clone https://github.com/gustavoandresfu/SciMapper.git
+cd SciMapper
+pip install -r requirements.txt
+```
+
+*(Optional)* Install as an editable command-line utility:
+```bash
+pip install -e .
+```
+
+---
+
+### 2. Initialize a Project
+
+Launch the interactive configuration wizard (prompts for domain preset, search queries, and project metadata):
+
+```bash
+python main.py --init
+# Or if installed via pip:
+scimapper --init
+```
+*(Or use `python main.py --init --quick` to generate the baseline template directly without prompts).*
+
+The generated `sms_config.json` is ready to inspect and refine:
+
+```json
+{
+  "project_metadata": {
+    "title": "Systematic Literature Review: Generative AI in Healthcare",
+    "short_name": "SciMapper_GenAI_Health",
+    "authors": ["Your Name"],
+    "temporal_window": [2019, 2026],
+    "output_dir": "./output"
+  },
+  "research_questions": {
+    "RQ1": "What foundational architectures and LLMs are predominantly adopted in clinical workflows?",
+    "RQ2": "What empirical metrics and validation benchmarks are reported for diagnostic accuracy?",
+    "RQ3": "What open challenges regarding hallucinations and ethical safety remain unaddressed?"
+  },
+  "search_strategy": {
+    "openalex_email": "researcher@university.edu",
+    "max_results_per_query": 100,
+    "queries": [
+      "\"generative ai\" AND \"clinical diagnostics\"",
+      "\"large language models\" AND \"medical decision support\""
+    ]
+  },
+  "nlp_tuning": {
+    "domain_preset": "biomedical_health",
+    "n_clusters": 5
+  }
+}
+```
+
+---
+
+### 3. Run the Automated Pipeline
+
+Execute the discovery phase:
+
+```bash
+python main.py --run-discovery --config sms_config.json
+```
+
+1. Review and refine taxonomic names in `output/discovered_topics.json`.
+2. Inspect the pre-selected seminal papers in `output/candidatos_discusion.csv`.
+3. Compile the interactive dashboard and Word document:
+
+```bash
+python main.py --compile --config sms_config.json
+python main.py --build-doc --config sms_config.json
+```
+
+Or execute everything end-to-end in a single run:
+
+```bash
+python main.py --all --config sms_config.json
+```
+
+---
+
+## 📊 Comparison with the State-of-the-Art
+
+| Capability | ASReview | Generic BERTopic | Manual Review Tools (Covidence / Rayyan) | **SciMapper** |
+| :--- | :---: | :---: | :---: | :---: |
+| **Academic Ingestion** | ❌ Manual (CSV/RIS) | ❌ None | ❌ Manual Import | **✅ Automated OpenAlex API** |
+| **Deduplication** | ⚠️ Basic | ❌ None | ⚠️ Exact match | **✅ Canonical DOI + Title Hash** |
+| **Semantic Representations** | ✅ Optional | ✅ Native | ❌ None | **✅ Pre-tuned Domain Transformers** |
+| **Topic Modeling ($c$-TF-IDF)** | ❌ Binary only | ✅ Yes | ❌ None | **✅ Mathematical $c$-TF-IDF** |
+| **Objective PRISMA Exclusion** | ❌ Human judgment | ❌ None | ❌ Human labeling | **✅ Algorithmic Outlier Clustering** |
+| **Cartography with Convex Hulls** | ❌ None | ⚠️ Scatter only | ❌ None | **✅ 2D Hulls + Centroids + Density** |
+| **Temporal CAGR Tracking** | ❌ None | ❌ Manual | ❌ None | **✅ Automated Hot-Topic Radar** |
+| **Automated Manuscript (.docx)** | ❌ None | ❌ None | ❌ Export tables | **✅ Formatted PRISMA Word Paper** |
+
+---
+
+## 🔬 Theoretical Foundations
+
+The framework's methodological and algorithmic foundations are grounded in established literature:
+
+1. **Topic Modeling & Semantic Representation:**  
+   Based on the modular architecture described by **Jay Alammar and Maarten Grootendorst** in *Hands-On Large Language Models* (O'Reilly, Chapter 5: *Text Clustering and Topic Modeling*), decoupling dense vector embeddings, dimensionality reduction, clustering, and class-based TF-IDF ($c$-TF-IDF).
+2. **Systematic Mapping Methodology:**  
+   Follows the seminal guidelines of **Petersen et al. (2008)** (*Systematic Mapping Studies in Software Engineering*) and **Kitchenham & Charters (2007)**.
+3. **Auditability & Reporting Standards:**  
+   Strict adherence to the **PRISMA 2020** statement (*Preferred Reporting Items for Systematic Reviews and Meta-Analyses*), ensuring transparent and reproducible record accounting.
+
+---
+
+## 📁 Repository Structure
+
+```text
+SciMapper/
+├── config_template.json    # Master template for project setup
+├── test_config.json        # Quick verification test suite
+├── main.py                 # Unified CLI orchestrator
+├── requirements.txt        # Python dependency specification
+├── pyproject.toml          # Standard packaging & CLI entry point (scimapper)
+├── CITATION.cff            # Academic citation metadata
+├── LICENSE                 # MIT Open Source License
+└── core/
+    ├── collector.py        # OpenAlex API ingestion & abstract reconstruction
+    ├── embedder.py         # Multi-domain Sentence-Transformer pipeline
+    ├── clusterer.py        # Dimensionality reduction (PCA+t-SNE) & c-TF-IDF
+    ├── selector.py         # Stratified seminal paper recommendation
+    ├── visualizer.py       # Plotly 2D interactive maps with Convex Hulls
+    └── doc_builder.py      # Automated PRISMA Word manuscript generation
+```
+
+---
+
+## 📖 Citation
+
+If you use **SciMapper** in your research, academic reviews, or dissertations, please cite it as:
+
+```bibtex
+@software{futo_scimapper_2026,
+  author       = {Gustavo Andrés Futo},
+  title        = {SciMapper: Automated Scientific Mapping & Systematic Literature Reviews with Dense Semantic Clustering and PRISMA 2020 Compliance},
+  year         = {2026},
+  publisher    = {GitHub},
+  journal      = {GitHub repository},
+  howpublished = {\url{https://github.com/gustavoandresfu/SciMapper}}
+}
+```
+
+---
+
+## 📄 License
+
+This project is licensed under the **MIT License** — see the [LICENSE](LICENSE) file for details.  
+Author: **Gustavo Andrés Futo** (2026).

scimapper-1.0.1.dist-info/RECORD

@@ -0,0 +1,15 @@
+core/__init__.py,sha256=E2r3eKcI1M1FfgvWdfFt55GrRc2ZA9sazMXLQEKU54k,262
+core/clusterer.py,sha256=H-rwlbPq3lCr8dHuP0Q42C0JSYOBd8KMH4e0hKxUcZ8,7499
+core/collector.py,sha256=AxfF6LE_ISmomKKUiyzjTat-uoppyVsRnQj6Bxi9r-8,6024
+core/config_template.json,sha256=jySu7Lc81gcGscAoQJ8olvBqYBnfrqFg6Y8Mjji6oNE,1627
+core/doc_builder.py,sha256=ykb1a03qwkOes8DFAlgw2H9OaOsoMa-HsPoE1huFJts,17108
+core/embedder.py,sha256=RDnuRokw3p75fn0DfjpF3T1cx4VjYO4lzJO09Cpk5xI,2998
+core/selector.py,sha256=VRRaKaieP1q1BQB1F9Ot54Amfuzz9umbJUoYr3J1arQ,6559
+core/utils.py,sha256=xthpQeghPjxW2JkYRE9qWa5kwefweUD0-rVEH2QQHwg,2866
+core/visualizer.py,sha256=92kC5AOVuChOH6ecquiM4nRfSjGvWMpqZyWbM4KyVhU,11848
+scimapper-1.0.1.dist-info/licenses/LICENSE,sha256=ml5HlDG7GbJ-jdXvUiYOaXv9rrrVuv6l1Ja_TUS92e8,1077
+scimapper-1.0.1.dist-info/METADATA,sha256=GvwG_di99DJ1QTdDMbxwTSjz2hvERF6dpL3bbf8zC5Q,12106
+scimapper-1.0.1.dist-info/WHEEL,sha256=mOdXR9C3wMVov7cSY8PY-XCzo3SmmWCHvGQr5csJoao,91
+scimapper-1.0.1.dist-info/entry_points.txt,sha256=2km7mWcGb_HIMK35IP1iZB8KlCe7NR4WJ6cMVxM9cEk,40
+scimapper-1.0.1.dist-info/top_level.txt,sha256=0ebp8k35dOTHVwbY3UcokRsjC7cWLN_iaJ70tIcmODU,5
+scimapper-1.0.1.dist-info/RECORD,,

scimapper-1.0.1.dist-info/WHEEL

@@ -0,0 +1,5 @@
+Wheel-Version: 1.0
+Generator: setuptools (84.2.1)
+Root-Is-Purelib: true
+Tag: py3-none-any
+

scimapper-1.0.1.dist-info/entry_points.txt

@@ -0,0 +1,2 @@
+[console_scripts]
+scimapper = main:main

scimapper-1.0.1.dist-info/licenses/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 Gustavo Andrés Futo
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

scimapper-1.0.1.dist-info/top_level.txt

@@ -0,0 +1 @@
+core