xsum_108_50000_25000_validation

This is a BERTopic model. BERTopic is a flexible and modular topic modeling framework that allows for the generation of easily interpretable topics from large datasets.

Usage

To use this model, please install BERTopic:

pip install -U bertopic

You can use the model as follows:

from bertopic import BERTopic
topic_model = BERTopic.load("KingKazma/xsum_108_50000_25000_validation")

topic_model.get_topic_info()

Topic overview

• Number of topics: 80
• Number of training documents: 11332

Click here for an overview of all topics.

Topic ID	Topic Keywords	Topic Frequency	Label
-1	said - mr - people - would - also	5	-1_said_mr_people_would
0	win - game - league - club - player	4532	0_win_game_league_club
1	police - court - said - mr - officer	2119	1_police_court_said_mr
2	world - sport - olympic - champion - gold	1037	2_world_sport_olympic_champion
3	bank - sale - growth - price - rate	604	3_bank_sale_growth_price
4	data - user - company - firm - mobile	225	4_data_user_company_firm
5	film - star - album - show - best	204	5_film_star_album_show
6	school - education - pupil - teacher - student	183	6_school_education_pupil_teacher
7	nhs - care - patient - hospital - health	107	7_nhs_care_patient_hospital
8	boko - haram - president - un - mr	103	8_boko_haram_president_un
9	labour - party - corbyn - ukip - mps	94	9_labour_party_corbyn_ukip
10	bird - specie - tree - ash - animal	82	10_bird_specie_tree_ash
11	northern - ireland - sinn - fin - dup	77	11_northern_ireland_sinn_fin
12	trump - mr - clinton - republican - president	73	12_trump_mr_clinton_republican
13	fire - blaze - smoke - scene - firefighter	73	13_fire_blaze_smoke_scene
14	water - flood - flooding - river - rain	67	14_water_flood_flooding_river
15	art - museum - artist - gallery - auction	63	15_art_museum_artist_gallery
16	transport - road - route - traffic - bridge	58	16_transport_road_route_traffic
17	ebola - virus - outbreak - vaccine - health	54	17_ebola_virus_outbreak_vaccine
18	syrian - syria - rebel - aleppo - iraq	53	18_syrian_syria_rebel_aleppo
19	race - hamilton - f1 - mercedes - rosberg	51	19_race_hamilton_f1_mercedes
20	welsh - wales - plaid - assembly - labour	51	20_welsh_wales_plaid_assembly
21	council - site - building - planning - centre	49	21_council_site_building_planning
22	rmt - rail - train - strike - service	47	22_rmt_rail_train_strike
23	ice - space - satellite - scientist - earth	47	23_ice_space_satellite_scientist
24	flight - plane - airport - aircraft - pilot	45	24_flight_plane_airport_aircraft
25	korea - north - china - korean - us	43	25_korea_north_china_korean
26	eu - uk - brexit - migration - immigration	43	26_eu_uk_brexit_migration
27	taliban - afghan - afghanistan - kabul - attack	43	27_taliban_afghan_afghanistan_kabul
28	hong - kong - china - chinese - liu	42	28_hong_kong_china_chinese
29	energy - wind - turbine - farm - electricity	42	29_energy_wind_turbine_farm
30	maduro - farc - president - venezuela - opposition	42	30_maduro_farc_president_venezuela
31	scottish - scotland - referendum - snp - independence	41	31_scottish_scotland_referendum_snp
32	cancer - risk - woman - study - diabetes	41	32_cancer_risk_woman_study
33	war - battle - royal - service - regiment	40	33_war_battle_royal_service
34	yn - ar - ei - yr - bod	40	34_yn_ar_ei_yr
35	space - mars - iss - astronaut - updated	39	35_space_mars_iss_astronaut
36	india - indias - delhi - hindu - indian	39	36_india_indias_delhi_hindu
37	russia - russian - ukraine - putin - president	38	37_russia_russian_ukraine_putin
38	tax - budget - chancellor - government - osborne	37	38_tax_budget_chancellor_government
39	coastguard - boat - lifeboat - rnli - rescue	32	39_coastguard_boat_lifeboat_rnli
40	elephant - ivory - zoo - animal - rhino	32	40_elephant_ivory_zoo_animal
41	pension - pay - wage - income - pot	32	41_pension_pay_wage_income
42	abortion - marriage - woman - samesex - gay	31	42_abortion_marriage_woman_samesex
43	paris - french - attack - france - jewish	29	43_paris_french_attack_france
44	dog - pet - animal - hare - police	26	44_dog_pet_animal_hare
45	unsupported - updated - playback - device - media	25	45_unsupported_updated_playback_device
46	pollution - waste - air - bag - bin	24	46_pollution_waste_air_bag
47	climate - carbon - emission - coal - gas	24	47_climate_carbon_emission_coal
48	mortgage - credit - lender - price - property	24	48_mortgage_credit_lender_price
49	ahmed - court - terrorism - police - naseer	24	49_ahmed_court_terrorism_police
50	train - driver - raib - incident - rail	23	50_train_driver_raib_incident
51	hoard - coin - museum - display - found	23	51_hoard_coin_museum_display
52	eu - trade - uk - market - brexit	23	52_eu_trade_uk_market
53	greece - greek - eurozone - debt - bailout	22	53_greece_greek_eurozone_debt
54	whale - shark - water - dolphin - fish	21	54_whale_shark_water_dolphin
55	steel - tata - industry - plant - uk	17	55_steel_tata_industry_plant
56	prince - duchess - duke - royal - princess	15	56_prince_duchess_duke_royal
57	camp - nazi - locsin - germany - extradition	15	57_camp_nazi_locsin_germany
58	migrant - refugee - border - turkey - greek	15	58_migrant_refugee_border_turkey
59	calais - migrant - camp - jungle - refugee	14	59_calais_migrant_camp_jungle
60	gun - violence - police - shooting - baltimore	14	60_gun_violence_police_shooting
61	rousseff - impeachment - senate - temer - petrobras	13	61_rousseff_impeachment_senate_temer
62	macron - le - germany - pen - macrons	13	62_macron_le_germany_pen
63	turkey - erdogan - turkish - turkeys - hdp	13	63_turkey_erdogan_turkish_turkeys
64	lodge - belfast - ballysillan - parade - paper	12	64_lodge_belfast_ballysillan_parade
65	castle - hull - staffin - crofters - house	11	65_castle_hull_staffin_crofters
66	food - cocoa - fairtrade - advertising - sale	10	66_food_cocoa_fairtrade_advertising
67	israel - hamas - israeli - gaza - palestinians	10	67_israel_hamas_israeli_gaza
68	bank - account - rbs - overdraft - note	10	68_bank_account_rbs_overdraft
69	runway - airport - heathrow - airports - flight	8	69_runway_airport_heathrow_airports
70	rescue - mountain - avalanche - climbing - nepal	7	70_rescue_mountain_avalanche_climbing
71	council - privatisation - deal - government - carmarthenshires	7	71_council_privatisation_deal_government
72	ruddy - inla - information - police - family	6	72_ruddy_inla_information_police
73	drug - prescribed - psychoactive - drugs - lyrica	6	73_drug_prescribed_psychoactive_drugs
74	kitty - book - author - publisher - prize	6	74_kitty_book_author_publisher
75	search - cabin - airways - plane - amsa	6	75_search_cabin_airways_plane
76	sterkel - alert - examined - detonated - rifle	6	76_sterkel_alert_examined_detonated
77	ambulance - service - aberglaslyn - called - inverclyde	5	77_ambulance_service_aberglaslyn_called
78	research - 3d - science - prof - vision	5	78_research_3d_science_prof

Training hyperparameters

• calculate_probabilities: True
• language: english
• low_memory: False
• min_topic_size: 10
• n_gram_range: (1, 1)
• nr_topics: None
• seed_topic_list: None
• top_n_words: 10
• verbose: False

Framework versions

• Numpy: 1.22.4
• HDBSCAN: 0.8.33
• UMAP: 0.5.3
• Pandas: 1.5.3
• Scikit-Learn: 1.2.2
• Sentence-transformers: 2.2.2
• Transformers: 4.31.0
• Numba: 0.57.1
• Plotly: 5.13.1
• Python: 3.10.12