Datasets:

bpalacios
/

chilean-political-signed-network

node_id string	canonical string	node_type string	degree int64	pos_degree float64	neg_degree float64	n_mentions int32	curated bool
N00196	Gobierno de Chile	institution	207,590	55,765	89,858	131,400	false
N00786	Gabriel Boric Font	person	108,299	37,085	41,655	80,694	false
N00001	Sebastián Piñera	person	97,910	28,345	42,700	79,048	true
N00005	Senado	institution	59,804	16,318	15,395	89,701	false
N00019	Donald Trump	person	51,907	9,637	31,936	29,482	false
N00010	José Antonio Kast	person	49,550	14,890	22,646	34,264	false
N00006	Michelle Bachelet	person	47,254	16,895	16,485	46,299	true
N00067	Cámara de Diputadas y Diputados	institution	45,348	13,954	14,253	44,322	false
N00021	Ministerio Público	institution	34,057	3,914	16,688	30,198	false
N00017	Estado	institution	32,742	10,858	13,864	25,243	true
N00008	Democracia Cristiana	party	30,869	11,987	11,852	44,227	true
N00003	Unión Demócrata Independiente	party	29,830	9,828	13,931	47,236	true
N00024	Renovación Nacional	party	26,125	10,033	10,199	48,511	false
N00700	Partido Comunista de Chile	party	26,044	9,630	12,331	44,493	false
N00018	Carabineros	institution	24,020	6,790	10,425	24,744	false
N00014	Chile Vamos	party	24,013	8,237	9,386	23,753	false
N00031	Contraloría	institution	22,772	1,604	9,287	19,496	false
N00016	Frente Amplio	party	22,290	8,324	9,883	27,596	false
N00028	Partido Socialista	party	21,685	8,755	8,037	40,064	false
N00034	Evelyn Matthei	person	20,921	6,438	10,058	13,078	false
N00036	Jeannette Jara	person	20,732	7,770	7,644	14,455	false
N00022	Partido Republicano	party	20,219	6,165	10,507	24,718	false
N00032	Camila Vallejo	person	17,312	5,576	7,431	12,733	false
N00049	Oposición	party	17,130	2,838	9,793	8,068	true
N00038	Nicolás Maduro	person	17,049	3,063	11,483	10,398	false
N00037	Carolina Tohá	person	16,347	5,242	6,239	11,682	false
N00025	Oficialismo	party	16,046	4,571	7,544	14,990	true
N00057	Daniel Jadue	person	14,939	3,447	8,666	9,012	true
N00039	Ministerio de Salud	institution	14,651	3,631	4,750	15,837	false
N00040	Corte Suprema	institution	14,639	2,265	7,422	12,395	false
N00046	Tribunal Constitucional	institution	14,306	1,798	6,674	11,372	false
N00064	Alejandro Guillier	person	14,154	5,011	5,634	8,684	true
N00013	PPD	party	13,918	6,201	4,324	27,133	false
N00068	Sebastián Sichel	person	12,834	3,763	5,710	7,896	true
N00058	Yasna Provoste	person	12,681	3,873	5,172	7,615	true
N00023	Presidente	person	12,664	3,667	4,583	20,318	true
N00069	Ministerio de Hacienda	institution	12,581	3,288	3,121	13,911	false
N00050	Mario Marcel	person	12,569	3,892	3,233	14,480	false
N00088	Ministerio de Educación	institution	12,214	3,067	5,427	9,698	false
N00052	Servel	institution	12,213	2,097	3,812	14,069	false
N00054	Giorgio Jackson	person	11,801	2,984	5,812	8,829	true
N00043	Parlamento	institution	11,746	2,892	4,312	11,541	false
N00073	Mario Desbordes	person	11,144	3,283	5,161	6,249	true
N00063	manuel monsalve	person	10,995	2,342	4,409	8,232	false
N00030	Nueva Mayoría	party	10,938	2,814	5,607	12,763	false
N13148	Estados Unidos	institution	10,739	1,995	6,643	52	true
N00079	Joe Biden	person	10,363	3,121	4,546	7,758	false
N00077	Izkia Siches	person	9,946	2,756	4,225	6,037	true
N00099	Manuel José Ossandón	person	9,334	2,024	5,104	6,295	false
N00055	Ministerio del Interior	institution	8,977	2,150	2,902	12,359	false
N05040	Chile	institution	8,730	3,705	3,139	455	false
N00141	Jair Bolsonaro	person	8,673	1,919	5,487	5,736	false
N00080	Andrés Chadwick	person	8,649	2,011	3,896	6,027	true
N00081	Ximena Rincón	person	8,348	2,554	3,296	5,859	true
N00045	Congreso Nacional	institution	8,232	2,354	2,069	9,178	false
N00093	Jaime Mañalich	person	8,227	1,664	4,321	5,839	true
N00071	Álvaro Elizalde	person	8,130	2,702	2,767	6,725	true
N00112	Evo Morales	person	7,878	1,573	4,851	4,129	false
N00091	Heraldo Muñoz	person	7,859	2,371	3,064	5,027	true
N00087	Karol Cariola	person	7,738	2,565	3,658	5,545	true
N00085	Carolina Goic	person	7,689	2,722	3,039	5,478	true
N00053	ricardo lagos	person	7,509	2,484	2,996	7,764	false
N00082	Enrique Paris	person	7,329	2,083	2,330	5,820	true
N00119	Maduro	person	7,284	1,118	5,190	3,954	false
N00047	ONU	org	6,913	1,851	2,395	14,329	false
N00094	Ignacio Briones	person	6,903	2,088	2,252	4,825	true
N00114	Unión Europea	org	6,875	1,967	2,866	5,616	false
N00103	Claudio Orrego	person	6,734	2,423	2,449	4,421	true
N00166	Javier Milei	person	6,710	2,067	3,322	3,084	false
N00090	Joaquín Lavín	person	6,662	1,820	2,267	5,229	true
N00041	Banco Central	institution	6,563	1,527	2,051	10,483	false
N00129	Francisco Chahuán	person	6,529	2,340	2,436	4,467	true
N00113	franco parisi	person	6,372	1,422	2,597	5,079	false
N00100	Juan Antonio Coloma	person	6,327	1,531	3,249	4,852	true
N00132	Matías Walker	person	6,212	2,231	2,453	4,131	true
N00121	Johannes Kaiser	person	6,106	1,242	3,360	4,072	true
N00246	Jacqueline Van Rysselberghe	person	6,056	1,629	3,145	3,817	true
N00108	Jaime Bellolio	person	5,990	1,844	2,592	4,342	true
N00138	Javier Macaya	person	5,910	1,608	2,886	3,924	true
N00234	Luiz Inácio Lula da Silva	person	5,877	2,159	2,648	4,323	false
N00075	Poder Judicial	institution	5,869	1,078	3,111	6,287	false
N00059	PDI	institution	5,858	1,560	1,728	7,343	false
N00111	Beatriz Sánchez	person	5,714	2,015	2,066	4,321	true
N00128	Marco Enríquez-Ominami	person	5,652	1,295	2,655	4,104	true
N00122	Carlos Montes	person	5,622	1,461	2,512	4,134	true
N00105	Codelco	institution	5,615	1,971	1,964	4,380	false
N00136	Servicio de Impuestos Internos	institution	5,613	816	2,685	7,040	false
N00172	Demócratas	party	5,599	1,815	2,549	5,464	false
N00149	Paula Narváez	person	5,572	2,382	1,744	3,520	true
N00115	Presidente de la República	person	5,509	1,380	1,487	4,681	true
N00160	Mauricio Macri	person	5,508	1,583	2,604	4,265	false
N00083	Mandatario	person	5,459	1,981	1,670	7,636	true
N00092	Augusto Pinochet	person	5,442	1,265	3,666	9,057	false
N00220	Dilma Rousseff	person	5,326	1,378	3,272	2,478	false
N00150	Marcela Cubillos	person	5,303	1,320	3,114	3,346	true
N00251	Convención Constitucional	institution	5,273	1,750	1,901	2,193	false
N00056	Apruebo Dignidad	party	5,270	2,520	1,557	7,611	false
N00137	Rodrigo Valdés	person	5,253	1,305	1,813	3,792	true
N00124	Hernán Larraín	person	5,247	1,567	2,144	4,036	true
N00123	Rodrigo Delgado	person	5,222	1,393	1,634	3,964	true

End of preview. Expand in Data Studio

Chilean Political Signed Network (2014–2026)

A signed directed network of Chilean political actors, extracted at scale from ~480k news articles spanning three governments (Bachelet II, Piñera II, Boric). Each edge is a polarized political act between two actors: actor_u → act_type → actor_v with a sign (+1 ally / −1 antagonist / 0 neutral).

480,002 articles → 239,684 connected actor nodes → 2,539,835 signed edges
Window: 2014–2026 · 3 administrations · two constitutional processes (2020, 2023)

⚠️ No news text is included. The source articles are copyrighted (CC-BY-NC), so this release contains only the extracted graph + metadata + article_ids — no article bodies, titles, or evidence quotes. article_id (md5(body)) lets holders of the licensed corpus join back to the source.

Files

File	Rows	Description
`nodes.parquet`	239,684	actors: `node_id`, `canonical`, `node_type`, `degree`, `pos_degree`, `neg_degree`, `n_mentions`, `curated`
`edges.parquet`	2,539,835	signed edges: `from_node_id`, `to_node_id`, `article_id`, `act_type`, `polarity`, `sign` (+1/−1/0), `issue`, `publish_date`, `period`
`articles_meta.parquet`	480,002	`article_id`, `source`, `publish_date`, `year`, `period` (no title/body)

Build the graph by joining edges.from_node_id / edges.to_node_id to nodes.node_id.

Statistics


Articles	480,002 (2014–2026)
Nodes (connected)	239,684
Signed edges	2,539,835
Polarity	42% negative · 30% positive · 28% neutral
Node types	person 140,603 · org 46,439 · institution 39,311 · party 6,390 · coalition 4,426 · movement 2,515

Top actors by signed degree: Gobierno de Chile, Gabriel Boric Font, Sebastián Piñera, Senado, José Antonio Kast, Michelle Bachelet, Donald Trump, …

How it was built (the text2SG pipeline)

A three-pass extraction, precision-first (a false edge pollutes the graph; a missed one only omits):

NER (entities). GLiNER (urchade/gliner_multi-v2.1, zero-shot, local, fp16) tags typed mentions: person / party / institution / coalition / movement / org. ~$0, deterministic.
Entity resolution (nodes). Surface forms are collapsed into canonical actors via token-blocked fuzzy clustering with cross-type and surname guards, then curated: a deterministic layer (exact-name + acronym dictionary, e.g. UDI ↔ Unión Demócrata Independiente) followed by a fan-out of LLM judges (6 parallel Sonnet adjudicators) that resolve the semantic grey zone (Ejecutivo / La Moneda → Gobierno de Chile; Fiscalía → Ministerio Público), strictly precision-first (never merges distinct people, never crosses types).
Relation extraction (signed edges). A tuned LLM extractor (gemini-2.5-flash-lite, given the resolved actors, abstaining without evidence) emits (actor_u → act_type → actor_v, polarity) grounded in an evidence quote. Validated at f0.5 ≈ 0.89 against a synthetic gold standard. The edge sign is the extracted polarity (positive=+1, negative=−1, neutral=0).

How it was validated (community detection + structural balance)

The network is validated not edge-by-edge but on its aggregate structure — the test that decides whether it is usable for analysis:

Community detection (Louvain on the positive/ally subgraph) reproduces the real Chilean political coalitions and their evolution across 12 years: Nueva Mayoría (Bachelet era) → Apruebo Dignidad (Boric era); the Communist Party bloc; and the split of the traditional right (Chile Vamos) from the republican right (Kast / Kaiser). Foreign actors cluster by ideology (Lula / Evo / Maduro with the left; Trump / Bolsonaro / Milei / Macri with the republican right).
Structural balance. 92% of the negative edge weight falls between communities (enemies are separated), consistent with balance theory — strong evidence the signed structure is reliable.

⚠️ Temporal role nodes. Actors like Gobierno de Chile, Oposición, Oficialismo, Presidente change referent at each change of government (e.g. Piñera → Boric, March 2022). Analyze their dyads per period/year, not aggregated over the whole window. Stable actors (people, parties) aggregate fine across all years.

Lineage

Built with the text2SG pipeline, evolved against an open synthetic benchmark:

text2signed-graph-gold — the synthetic gold standard (287 articles, 914 signed relations).
text2graph-evolve — evolutionary engine that optimized the extractor against that gold (precision-first, f0.5 fitness).
chilean-political-dataset-signed-networks — this dataset: the champion applied at scale to 2014–2026.

Citation

By Benjamín Palacios. Part of ongoing work on signed political cleavage networks in Chile.

Downloads last month: 38

Collection including bpalacios/chilean-political-signed-network

Chilean Political Signed Networks

Collection

2 items • Updated Jun 24