Take Home Exercise 3

Data preparation

load libraries.

pacman::p_load(jsonlite, tidygraph, ggraph, igraph,
               visNetwork, graphlayouts, ggforce, 
               skimr, tidytext, tidyverse, widyr, kableExtra, RColorBrewer)

Read MC3 data from Json file.

mc3_data <- fromJSON("data/MC3/MC3.json")

Extracting edges

mc3_edges <- as_tibble(mc3_data$links) %>% 
  distinct() %>%
  mutate(source = as.character(source),
         target = as.character(target),
         type = as.character(type)) %>%
  group_by(source, target, type) %>%
    summarise(weights = n()) %>%
  filter(source!=target) %>%
  ungroup()

Extracting nodes. “product_services” column value with “character(0)” are replaced with “unknown”. “revenue_omu” with NA are replace with numerical value of ‘0’

mc3_nodes <- as_tibble(mc3_data$nodes) %>%
  mutate(country = as.character(country),
         id = as.character(id),
         product_services = ifelse(product_services == "character(0)", "unknown", as.character(product_services)),
         revenue_omu = coalesce(as.numeric(as.character(revenue_omu)), 0),
         type = as.character(type)) %>%
  select(id, country, type, revenue_omu, product_services)

Initial Network Visualisation

Show the code

id1 <- mc3_edges %>%
  select(source) %>%
  rename(id = source)
id2 <- mc3_edges %>%
  select(target) %>%
  rename(id = target)
mc3_nodes1 <- rbind(id1, id2) %>%
  distinct() %>%
  left_join(mc3_nodes,
            unmatched = "drop")

mc3_graph <- tbl_graph(nodes = mc3_nodes1,
                       edges = mc3_edges,
                       directed = FALSE) %>%
  mutate(betweenness_centrality = centrality_betweenness(),
         closeness_centrality = centrality_closeness())

mc3_graph %>%
  filter(betweenness_centrality >= 100000) %>%
ggraph(layout = "fr") +
  geom_edge_link(aes(alpha=0.5)) +
  geom_node_point(aes(
    size = betweenness_centrality,
    colors = "lightblue",
    alpha = 0.5)) +
  scale_size_continuous(range=c(1,10))+
  theme_graph()

Pairwise_similarity on product_services (business)

Tokenisation to product and services of each business.Regroup back after clean up at token level and assign a code ‘biz_code’ to every distinct services. Total nodes after clean up = 3894 (org_id_code)

#Original id, service and code 3894 (un group)
org_id_code <- mc3_nodes %>%
  rename(company = id, business = product_services) %>%
  unnest_tokens(word, business) %>%
  anti_join(stop_words, by = "word") %>%
  filter(word != 'unknown', word != 'related',word != 'products',word != 'product') %>%
  group_by(company) %>%
  arrange(company,word) %>%
  summarise(business = paste(unique(word), collapse = " ")) %>%
  mutate(biz_code = dense_rank(business )) %>%
  ungroup()

head(org_id_code, 100) %>%
  kbl(caption = "Orginal nodes after clean up") %>%
  kable_paper("hover", full_width = F) %>%
  scroll_box(width = "100%", height = "500px")

Orginal nodes after clean up
company	business	biz_code
1 AS Marine sanctuary	accessories beads bits craft crafty diy doll embellishment funfoam kits paper ribbons scrapbook shapes stencils styrofoam tools wood	182
1 and Sagl Forwading	logistics solutions total	2881
2 Flounder ОАО Consultants	beverages canned condiment dried drinks eat fish food frozen leaves meat mix noodles pickled pouches preserved ready rice sauce seafood snacks spices tea	1439
2 Limited Liability Company	aquatic canning manufacturing processing seafood	990
3 Limited Liability Company	apparel discount fabrics fibres garments greige home market segments textiles yarns	885
3 Ltd. Liability Co Transit	european food specialty	2475
3 SE & Son's	anodes chemicals industry plating worldwide	848
4 Limited Liability Company Seaport	cold freezing providing services storage	2055
5 Flounder S.p.A. Transport	engineering ferrous industry iron metals sector services steel	2424
5 Limited Liability Company	3 adhesive adhesives agents chemicals combination development environmentally friendly glue hot instant involved melt performance polymer production sales special specialises trade	39
7 GmbH & Co. KG	blades cast cutlery cutting edge flatware hand handsaws including iron kitchen machines manufacture metal nonpowered nonprecious pans plated pots precious sawing skillets tools type types utensils	1482
7 S.A. de C.V. Shipping	bar carbon reinforcing rounds squares steel	1303
8 S.A. de C.V.	antibodies biopharmaceuticals proteins recombinant	859
8 SE Marine life	gloves niche	2776
9 Charter Boat Plc Overseas	fish	2542
9 GmbH & Co. KG Solutions	accessories conveyors fire range valves	214
9 Shark S.A. de C.V. Brothers	ang bean bing bowls dzu flour geum ice jelly juk kan ki paste pat porridge processed red rice shaved su sweet yo	809
Acevedo-Hall	snapchat spectacles	3042
Acosta Inc	ability accessories allowed apparel aspects assortment balance brands capitalize category comprising core customers decades derive design designing designs development distribution dynamics excelled exclusivity experience fashion focuses footwear forward global globe handbags house industry international kids labels maintain manages manufacturing market men's operations presence prevailing principles private producing production provide reputation responsible round sellers service specializes strong styles trends variety wide widespread women	70
Acosta, Proctor and Meyers	beef chicken cream dairy deep diced fish freeze frozen ice jams jellies juices lamb meats pork seafood sliced vegetables yogurt	1407
Acosta, White and Patel	chemicals industrial inorganic	1951
Adair - SE Transport	bags leather	1245
Adair Ltd. Liability Co	aftermarket automotive covers customizing electronic enclosures fobs frames hitch including key license manufacturing plastic plate trailer	507
Adair S.A. de C.V.	cephalopod cuttlefish frozen octopus soft squid	1903
Adams Group	fish range seafood	2585
Adams-Hicks	clothing footwear	2012
Adkins LLC	frozen seafood	2734
Adkins, Garcia and Perkins	companies distributes electric electronic frequency hydraulic lines power stations thermal transformer	2091
Adkins, Williamson and Daniels	boats landcraft sailboards sails	1523
Adkins-Villanueva	harnesses wire	2798
Adriatic Nautical CJSC Holdings	act agent business cargo class clearing consigned customs facilities forwarding freight infrastructure insurance land logistics marine ministry ocean operating property sea services shipment tourism transport transportation type voyage warehousing	287
Adriatic Seabass LC Transportation	brokerage complete custom forwarding freight logistic providing service shipping solution world	1634
Adriatic Catch Ges.m.b.H. Forwading	food	2619
Adriatic Catch Ges.m.b.H. Marine biology	consumer	2167
Adriatic Catch Ltd. Liability Co	machine tools	2886
Adriatic Catch S.A. de C.V.	analysis technical testing	789
Adriatic Catch SRL Seal	construction dedicated engineering key plants plastics unit	2154
Adriatic Mackerel Ges.m.b.H. Worldwide	boats complete elements laminate models moulds operates produces subcontractor	1520
Adriatic Mackerel N.V. Transport	appetizers aromatic asparaguses caviars chilean clams cockles cods crabs cuttlefishes filets fish fruits green hearts herbs mackerel maizes mangoes mussels octopuses oysters packaged palm papayas peppers pineapples prawns rambutans razors sauce sauces seafood seasonings spices spicy squids stuffed sweet truffles tunas vegetables winkles	902
Adriatic Squid AS Holdings	canned seafood	1775
Adriatic Squid Limited Liability Company Freight	cured fish fresh frozen seafoods	2275
Adriatic Squid Ltd. Liability Co	finfish footwear including seafood shellfish shrimp	2535
Adriatic Tuna GmbH & Co. KG	forwarding freight integrated operation warehouse	2686
Adriatic Tuna Seabass BV Transit	petfood tuna	2975
Agua N.V. Marine life	2 pole turbogenerators	30
Aguamarina Beach AB Industrial	artichokes beverages bread cheese coffee crumbs desserts gnocchi hearts mediterranean oil olive olives palm pasta pickled rice seafood soups specialties spices tomatoes tortellini vinegar	1056
Aguamarina Limited Liability Company	fresh frozen seafood	2720
Aguamarina S.A. de C.V.	compounding equipment extruding measuring pelletizing	2131
Aguilar LLC	motor vehicle	2941
Aguilar, Smith and Perez	cream fish food frozen fruit ice meals pastry ready seafood vegetables	2254
Albertine Rift Company Enterprises	footwear	2669
Albertine Rift NV Family	alaska chum cod cucumber fish hake pc pink pollock pop red salmon sea shellfish	659
Alexander, Sullivan and Grant	eco equipment friendly industry machinery nonferrous plant steel supply	2393
Ali, Barber and French	decorative materials wood	2305
Ali, Castro and Cruz	applications cement chemical commercial communication complexes distribution energy fertilizer furnace including independent industrial locomotive media metering meters mills petro plants power purpose rectifier residential solutions special static steel support traction transformers utilities	954
Allen Ltd	bass fresh sea tuna yellowfin	1355
Allen PLC	fish materials plastics resins seafood	2574
Allen, Thomas and Jackson	engines gearboxes vehicles	2428
Allen-Brown	materials plastic resins	2908
Allen-Guzman	accessories apparel clothing	146
Allen-Robles	accessories equipment inspection machine precision surface tooling tools	224
Alvarado, Jensen and Brown	acid based car cement cements coating coatings coats deck decorative duty external finishes floor flooring hardeners heavy levelling linings mortars park potable resistant skim tank water waterproofing	262
Alvarado-Oliver	adhesive coating paint	408
Alvarado-Sanchez	anti compounds emulsions fluids foam greases miscellaneous silicone	855
Alvarez, Harrell and Zimmerman	food	2619
Alvarez-Cabrera	carbon coils flat painted ppgi pre sheets steel strips tubes welded	1809
Alvarez-Rogers	aerosols paint putties touch up's	499
Amerigo River S.p.A. Express	assistance business change coaching development implementation improvement management operational organizational services skills strategy support technology	1115
Amur Estuary Pic Cargo	air compressor hand hardware hydraulics pressure tools trailer washers welding	609
Amur GmbH & Co. KG	foodstuff gelatines ingredients	2666
Amur Holdings Coral Reef	food merchandise	2645
Ancla Azul SRL Coral	plastic	2978
Ancla Azul ОАО Holdings	accessories alterations basic fabric fabricating fabrics fiber fibers finishing food frozen hemming knitting lengthening letting logistics marketing notions piece preparation refrigerated services shortening taking thread weaving yarns	126
Ancla Marina Limited Liability Company	leather shoes	2865
Ancla Marina Ltd. Liability Co Harbor	courier providing services specializes trucking	2235
Ancla Marina N.V. Brothers	food	2619
Ancla de Oro & AS Carriers	meat	2914
Ancla de Oro AB	automotive industry systems wiring	1201
Ancla de Oro Pelican	alaska blocks bulk claws clusters cod cooked crab crawfish dungeness fillet fillets gutted head iqf ivp king legs pacific pollock portions red salmon shatterpacked snow sockeye tilapia	657
Ancla del Este AG	bioactive collagen fish ovalbumin peptide soybean	1461
Ancla del Este LC	collagen gelatin supplements	2062
Ancla del Este LLC	fish seafoods yarns	2593
Ancla del Este Sagl	bottom catfish clams cod crab exotics farmed fish fishes flounder halibuts oysters pollock rockfish salmon shellfish shrimp sole tilapia tuna whiting	1570
Ancla del Este Sp Fish	distribution fishing operation seafood vessels	2345
Ancla del Este Starfish	adults basic blended color control cotton deliver dyeing elevated fabric fabrications finishing garments hand juvy line providing seamed shrinkage technology toddler top tubular youth	469
Ancla del Este ОАО	seafood	3023
Ancla del Mar Express	appliances applicator auto beauty ceramic components cosmetic design equipment filling fine hot lamination machine manufacturing melt range specialises ultrasonic wide	917
Ancla del Mar Pic Worldwide	fish seafoods	2592
Andaman Jetty S.A. de C.V. Marine ecology	accessories extensions eyelash eyelashes false focuses strip	226
Anderson Group	arc auxiliary casting ccm continuous eaf electric food furnaces ladle lf machines melt mills refining rolling shop systems	1016
Anderson Ltd	business diving gas maritime measuring personal protection services solutions system technologies technology transport	1686
Anderson, Gross and Becker	accessories home office personal	235
Anderson, Russell and Allen	food ingredients	2636
Anderson, Tucker and Williams	valve	3060
Anderson-Marsh	adhesives chemicals industrial sealants surface treatments	442
Anderson-Nash	casting iron	1864
Anderson-Olson	fish	2542
Andhra Pradesh BV Marine life	based commerce wholesaler	1333
Andhra Pradesh CJSC Merchants	automatic cutting systems	1164
Andhra Pradesh Corporation Holdings	accessories bags diving fish fishing heads hooks inshore line lure monofilament skirts tackle terminal trolling	172

Performed group by biz_code. Nodes by unique biz_code = 3064 (grouped_biz_code)

#grouped_biz_code (3064 unique service and code) lost id relationship
grouped_biz_code <- org_id_code %>%
  group_by(biz_code) %>%
  summarise(business = paste(unique(business), collapse = " "), biz_code) %>%
  distinct()

head(grouped_biz_code, 100) %>%
  kbl(caption = "Unique bussiness with assigned code") %>%
  kable_paper("hover", full_width = F) %>%
  scroll_box(width = "100%", height = "500px")

Unique bussiness with assigned code
biz_code	business
1	0.5a 1.5a 1a 24w 2a 5v 90 adapter charger desktop driver gaming interchangeable keyboard laptop led lighting machine massage medical metal mouse power usb
2	000 1872 27 4 alligator aluminium analysers analyzers balers based bespoke business cable car casting complete container de design equipment falls ferrous freelance granulators history hydraulic including ingot loaders machinery machines main manufacturer material mcintyre metal nottingham pollution processing range reconditioned recycling researcher sales scrap services shears source support system systems tilters unloaders worldwide
3	0302.91 0302.99 0304 alaska chalcogramma chilled edible excluding fillets fish fresh meat offal pollack subheadings theragra
4	0302.91 0302.99 0304 australasicus chilled edible excluding fillets fish fresh japonicus mackerel meat offal scomber scombrus subheadings
5	0302.91 0302.99 0304 brisling chilled edible excluding fillets fish fresh meat offal pilchardus sardina sardinella sardines sardinops spp sprats sprattus subheadings
6	0303.91 0303.99 0304 albacares edible excluding fillets fish frozen meat offal subheadings thunnus tunas yellowfin
7	1 20 60 alfred bacteriology collection controls devices esr hb latex lc line modern penok penol pn roller sidecar swab test tests
8	1 250 27 aluminum auxiliary cleaning complex components container cranes device electrical engineered equipment finished forklift grade handling hydraulic injection lifting loaders machines manipulators materials molding pallets platforms roads rollers scissors season semi sidewalks steel structures tracks trailer trolley winter
9	1 4 5 6 7 automatic auxiliary class cvt fwd gearbox hybrid liter medium mini rwd speed transmission vehicles
10	1 800 alloy cast gram grams precision weight zinc
11	1 accessories art arts backpacks binders board boards book books calculators calendars cards clips coloring composition construction cooler correction covers crafts crayons desk document dry envelopes erase erasers fashion fatbooks foam glue highlighters holders home including index lunchbags mailing markers memo notebooks offers pads paper pencil pencils pens pins plastic portfolios poster project punches push range rulers scissors sharpeners staplers subject supplies tape totes wide
12	1 bound cream crystal drink drinking fish gum ice jelly konjac milk papaya pudding refined seafood soft sorbet stabilizer suspension thickening water wine yoghurt
13	10 100 acetyl acid alpisarin ammonium arteether artemether artemisinin artesunate baccatin beta bp bromide butyl camptothecin coenzyme colchicine dab de digoxin dihcl enoxolone ep forskohlin glycyrrhetinic glycyrrhizinate hcl herpes hydrobromide hyoscine hyoscyamine jp machinery mag manufacturing metalworking methoxsalen nicotine noscapine ohimbine organizations outsourcing paclitaxel piperine podophyllum private public quinidine quinine reserpine resin salts sennosides service solutions sulphate tartarate thiocolchicoside u.s.s.r.p usp xanthotoxin
14	100 280 320 4 80 acid alka bloom cold concentrations extraction flavoured food gelatine gelatines gold grade granulometries hydrolysed instant leaf line liquid lumps mesh pharmaceutical platinum silver soluble technical type wine
15	100 anchovies black green manzanilla olives peppers percent recipes spanish stuffed
16	100,000 200,000 60,000 80,000 africa australia bulk capacities cape cargo cargoes cement chemical coal continent conventional countries crude dangerous dry dwt east fertilizers fleet gas gasoline grain guatemala handymax hbi heavy including india iron lift liner methanol middle naphtha north offers oil oils operates ore p.g pacific palm panamax petcoke ports ranging salt sand scrap services silica sized south suezmax sugar sulphur tankers tramp transportation transporting vessels vlccs
17	110v 220v 3 air chilled coils commercial concealed condensing ducted dx evaporator fan handlers include mount portable residential series ton unit units universal wall water window
18	12 84 automotive card cards convenience countertop custom display displays drug fixtures floor greeting grocery hold hundreds literature mass merchandise merchandising pockets racks sorts specific store stores wire
19	12 axle bearing bridge components gear heavy internal mechanical reduction ring seals speed steyr transmission truck wheel
20	120 370ml 580ml 6 9 asparagus canned diamir fa fruits hood light marmelades meals mushrooms ol olive pepper piq ready seafood tuna vegetables ventresca
21	13 ac chassis condensing consoles custom freelance heat horizontal list low packaged profile ptac pumps remote replacement researcher seer source stacked unit units vertical
22	150 advanced annual approvals assembly automatic ce certificates completed erp halogen lamps led lights line million output pieces production reach series sets type
23	16 braided complete cord cordage cords cotton diamond line nylon polyester polyethylene polypropylene rayon rope ropes round smooth solid specializing synthetics twine twisted
24	1604 fish heading minced n.e.c pieces preparations prepared preserved
25	1974 acquired affected assembly bar businesses closer crisis development direction distribution evolutions faster feeder field fine future hydraulic hydrobar ingenious inventing invests jura leader loaders machine machining makers markets mechanical medium peripherals philippe position precision publicly pushing reached region regional respond scheurer sized storskogen swedish technician tool traded tuned watch worldwide
26	1999 catalogs children's clothing launched mail men's retailer sells site wear web women's
27	2 200 24 atmos black box cans care cleaning collab crates crep cure era eraser flat footwear kicks kilates kit laces limited mark ml multi nba nice pack pen pill pre protect quantities refill round shoe sneaker spray team treated trees tube ultimate white wipes â
28	2 3 360 3ds 4 accessories action adventure apparel blu board cards console consoles content digital ds dvds family fighting figurines games guides hardware headsets keyboards mice music nintendo online pc playing playstation portable posters puzzle racing ray replicas role shooter simulation sport statues strategy tickets toys trading video vita wii xbox
29	2 3 caps containers cutting easy lithography metal piece plate traditional twist varnishing
30	2 pole turbogenerators
31	200 blanca canada canadians check companies company employs fisheries fishing freelance head job john largest liability limited management nataagnaq newfoundland nunavut office offshore opportunities playa researcher source st
32	2202 3303 411 441 551 553 fume metal powder silica silicon
33	24 ball carat correction erasers gel items lead pens promotional refills scales sharpeners stationery
34	25 biaxially bopp cast cpp degradable film films gba gbc gbp gcl gcm gcm25m gec12m gec12s green guolene metallized oriented polypropylene
35	250 acid advisory alloyed alloys automotive cast casting chemical construction development finishing fittings heat highly including industries industry iron kilogram low material offers post process prototype pump purpose range research resistant service special steel surface treatment und valves weight
36	3 3.0 accessories adapters card converter devices docking hub lan media multimedia ports reader sharing station stations thunderbolt travel usb
37	3 99 acid agro agrochemical aluminium bases bulk chloral chloride chloroacetylechloride drugs dye ethyl fast formate formic formulations fungicides herbicides hexahydrate including intermediates methyl minimum nitrophthalic nitrophthalonitrile percent pesticides pharmaceuticals
38	3 acids botanical extracts fatty omega probiotics
39	3 adhesive adhesives agents chemicals combination development environmentally friendly glue hot instant involved melt performance polymer production sales special specialises trade
40	3 cod fish liver oil omega shark
41	350 400 6 750 approximately boutiques brand departmental exclusive including john lifestyle multi operates outlets players stores wills
42	3501 animal casein coloured derivatives excluding gelatin glues heading including isinglass origin rectangular sheets surface
43	3d adstra arm cabinetry cabinets cart chairs charting clinics compressors computers cuspidors delivery den dental equipment equus film furniture imaging instrument intraoral lighting management microscopes mobile nitrous operatory oxide packages panoramic portable practice precision processors pumps ray reception refurbished seiler software sterilizers stools systems tech units vacuum washer
44	3d anti applications climate components control crystal decorations diffuser display displays films flat graphic hygiene incorporating interactive liquid manufacturing microbial mould optical panel panels precision printing properties screen sector suitable systems touch
45	4pl air brokerage bulk cargo commerce contract controlled customs distribution engineering forwarding freight include information insurance international land load logistics management manufacturing multimodal national ocean packing pallet preparation reverse services shared shipments solutions stock temperature transport transportation warehousing
46	5 automatic axis bridge changers horizontal machines milling pallet production speed tapping vertical
47	5 bacon biscuits chewing convenience corn crocs dishes fish ham palette pasta pizza poultry ready sandwiches sausages seafood serve snackies sticks strips succulent tasty wide
48	5 beans bio black borlotti butter chapped cherry chick chilly crushed double eye flavorings fruits garlic green herbs hill lentils mix onion paste peas peeled pizza sauce tomato tomatoes vegetables velvety white
49	500 chain company customers delivers germany largest migros parent party stores supermarket switzerland switzerland's
50	500w amplifiers cim commutation components controllers core customer dc drive electronic encoders flat gears iron ironless mechanical mim motors output planetary position power resolvers sensors servo solutions special specific spur tachos tech winding
51	6 backpack bag business compu computer cooler messenger polyester portable sling sports tote trail travel
52	600 asprin bayer bone calcium capsules care cats claw devil's diet dogs electrolyte formulas gemini herbal joint liposafe minerals replenishment selenium supplement tablets women's yeast
53	63 77 83 92 action female laminate lock sv taper tilt valves vertical
54	8 air apl arabia atlantic ccl cement container continental freelance freight line logistics mixer ocean pacific pal project researcher source units
55	8 balls beer cubes drives effects environment flat items magic memory picture prism protection recycle rubik's sliding tangles usb
56	80 camp developmental disabilities electric freelance gas ground heaters heating manufacturer nonprofit ofsolar pool pools private researcher solar source summer swimming system systems thermal
57	8903 boats canoes heading inflatable n.e.c pleasure rowing sports vessels yachts
58	900 adapters cables cords coupler couplers distribution hybrid indoor lc lszh mt panels patch producing pvc range rated riser rj specialized st standard style wide
59	a.o automation chinese class electro european leading manufacturers offering represents specialises suppliers technology
60	a5 airpen blue computer diagrams electronic format handwritten hundred ink kga memory normal notebook notes pages paper pen personal sketches storage stores uploaded ventures water writes
61	aaa albacore arkshell bighead chuka clam crab croaker cut escolar fillet fish flying frozen fulvescen ground half hamachi hiyashi ika items ivp izumidai korean loin mackerel madako meat mussel nobashi octopus oyster pike poke pollack ribbon saku sansai sea seafood seaweed shell shisamo slice snail snow squid squirt steamed sushi tako tiny tobiko tuna wakame yellow
62	aalesund agency airfreight airlines belgium break broker brokerage bulk bunker cargo cargoes chartering chilled consignments consolidation container containers crew customs declaration delivery destinations direct discharging door dry eastern europe exports extended france frozen germany handle handles handling imports including items kinds land lay loading lufthavn netherlands norway offers oil operates operating planning port ports project racks sea services ship shipchandler shipping skutvika stevedoring storage stores terminal transfer transport transportation transshipment truck trucking vessels vigra worldwide
63	aasicp analysis australian certified coal freelance materials minerals reference researcher source standards supplier
64	abalone anchovy blowfish bluefish butter capelin catfish caviar clam cockle conch crab crayfish crustacean cucumber cuttlefish dogfish echinoderms eel fish flounder flying grouper haddock kingfish lobster lumpfish mollusks mussel octopus oyster periwinkle prawns roe salmon scallop sea shad shellfish shrimp snail squid urchin
65	abalone cactus dry ezo fish flounder food foods frozen ingredients scallop seasonal yellowtail
66	abalone hake pink prawns seafoods
67	abalone mollusc preparations prepared preserved
68	abc advanced aim aluminum applications board bonded circuit class copper development electronic foils freelance interconnect materials performance printed production researcher source world
69	abi ach automated broker cargo clearing consular credit destinations documentation door export freelance house insurance interface letters researcher service shipment source
70	ability accessories allowed apparel aspects assortment balance brands capitalize category comprising core customers decades derive design designing designs development distribution dynamics excelled exclusivity experience fashion focuses footwear forward global globe handbags house industry international kids labels maintain manages manufacturing market men's operations presence prevailing principles private producing production provide reputation responsible round sellers service specializes strong styles trends variety wide widespread women
71	abrasive abrasives accessories air bases batteries bead bench benches blanks books brushes buffs bur burs busch carbide carving casting chairs cleaners cleaning cloth compounds compressors cratex crucibles cup cutters decorative diamond display drawing dressing drills electroplating engraving equipment felts files filters finishes flap flasks flex frames gemstones grain grinders hammers handpieces hears hoses imaging inside investing kits laps lighting machines mallets mandrels measuring mills nippers packaging photography pins pliers polishers polishing power range reamers rifflers ring rolling safety sandblasting saws scrapers scribers setting shafts shop soldering solutions specialises stamps stands stone stones stringing supplies supplying taps tools torches tumblers tweezers ultrasonic vacuums videos watch wheel wheels wide wildfowl
72	abrasive aluminum ceramic fused industries line oxides refractory vast
73	abrasive ash concrete control equipment handling material multi pollutant systems
74	abrasive brakes building cnc controlled cutting guillotine including machines offers plate press presses pressure pumps robotized routing shears ship systems tools waterjet waterjets
75	abrasives agglomerates brushes ceramic cutting epoxy equipment granites marbles mastics surfaces systems technical tools treatments
76	abrasives applications based expendable quality recyclable slag specialty variety wide
77	abrasives bars bits boring boxes broaches cabinets carbide chucks collets coolant counterboxes countersinks cutters cutting drill fasteners files grinding handtools heads holders indexable instruments lathe measuring mills precision reamers rotary saws shop supplies systems tapping tool tooling tools vises wheels
78	abrasives cabinets canal cements dental equipment hand laboratory ray resins root stones units waxes
79	abs accessories aerobic amounts apparel bikes boxing catalog dry equipment exercise fitness furnishings groceries hardware home house kits lifting limited line lines machines mail massagers media merchandise pilates steppers steps strength television trainers training trampolines treadmills vibration wares weight yoga
80	abs acrylonitrile butadiene chloride hips impact nylons pa pc pe pes pet plastic poly polyamides polycarbonate polyester polyethylene polypropylene polystyrene polyurethanes polyvinylidene pp ps pu pvc pvdc saran styrene terephthalate vinyl
81	absorbent ammonium anhydrous antimony bicarbonate bone caffeine chloride diphyenyl ethyl glue guanidine gum lead mono oclansorb oil phosphate phosphorus potassium red rosin sodium sulphate tartrate trisodium xanthate zirconium
82	absorber applications automation bike components compressor comprise comprising connecting consumer cum electronic engine forging heat hubs include including industrial insulin integration lens machining medical metal optical processing professional rail rings services shock sleeves suspension syringe system timing treatment variable
83	absorbers added chemicals coatings finishes flame innovative management moisture oil release repellents retardants soil stain uv water
84	absorbers anti articles bar bearing bearings calibrated cast components connecting conservatories consist cross differential doors doorways drive elements engineering extraction extrusion facades finger fixing food fresh gaiters gaskets gearbox guard hydraulic include including industrial industry intermediate limiters linings membrane moulds mountings movement pipes profiles protection rods roll rubber sealing shaft shock slitted steering stop struts system technical technology transmission waste water white windows zsb
85	absorbing active adjustability adjustable advanced aftermarket assemblies columns components comprising drive electric electronics energy equipment final front half hydraulic including intermediate joints kits modules original overlay power propeller rake rear services shaft shafts solutions steering systems technology tilt torque touch wheel
86	absorption ac anti apparel breathable coating cpu cut dirt durable dyed dyeing embossing fabric fabrics fashion fiber flame form furnishing germs home including industrial laminating lamination mesh mildew nylon olefin outdoor padded polyester pressing printing process proof pu pvc recycle recycled repellent short special spun textile textiles tricot uv vinyl water yarn yarns
87	ac accessories arc beam billets blanks blooms building casting continuous dc degassing electrical equipment fibers furnaces gift household indoor industrial ladle machines materials minimills outdoor plants rounds sewing stationery supplies threads vacuum vd vod yarns
88	ac automotive brushless cleaner dc gear include induction motors offers pole range shade universal vacuum
89	ac dc freelance gear hydraulic manifolds motors power pumps remotes researcher reservoirs source units
90	acacia almond blossom calluna chestnut eucalyptus forest heather honey lavender lemon linden mandarin mountain orange organic origins polyflora rape rosemary spanish stoechas sunflower thyme tree types vitex
91	accelerated business consultancy freelance industrial joint performance researcher service source supply testing tires tosupport venture wheels
92	accelerator adhesive adhesives anaerobic bottles curing cyanoacrylate dispensing dosing epoxy equipment floodlight glue lamps led linop medical methacrylate methyl packaging primer remover solutions tips uv
93	accent accents americana bow bridal check christmas craft decorative floral gift holiday industries items lace manufacture metallic offers patrick polypropylene portfolio ribbon ribbons sale satin seasonal sheers special spring st striped supplies tulle unwired valentine velvet waterproof wedding wired wrapping
94	accent commotions cones cookie cool cream creamy desserts drink drinks duetto energy frozen fruit fusion gelati ice iced including light moos nature's premium sandwiches sherbet sundae tea treats venice yogurt
95	access acute aesculap anesthesia apparatus care chronic clinical closure dialysis disinfection divisions endoscopes fields frequency gloves hospital hygiene incontinence infusion injections instruments intensive intravenous invasion management market medical medicine minimal nutrition offers operates opm parental patient services solutions stoma surgical therapy treatment unit urology wound wounds
96	access ade air architectural benchmark boards cable centre components containers control controlled data door drainage ductwork energy engineered environmental environments factory faã floor floors frame framing harvesting heat height hot hvac insulated insulation integrated management panels piping pollution pumps rainwater renewable roof roofs safety solar solutions source steel structural sustainable systems telementry timber underfloor wall walls water wire
97	access arm articulated capacity compact equipment forklift forklifts front including industrial loaders maniaccess manihoe manilec maniloader maniscopic manitou manitransit masted mounted pallet platforms rear rotating rough scissor semi shovel stackers telescopic terrain truck trucks vertical warehouse
98	accessibility assistance buses city coaches mini services support systems technical
99	accessories acoustic adhesives automotive covering energy engineered fire foams industrial insulation markets metallic offers pvc range safety systems technical wind
100	accessories acrylic appliques beaded beads clay cloisonne collars corsages crochet embroidered fancy fashion fringes garments gemstones glass hand jeans motifs natural scarves shells silk skirts tops trimmings

Redo Tokenisation base on grouped_biz_code and perform pairwise_similarity business

#Tokenisation of service words
grouped_service_code_words <- grouped_biz_code %>%
unnest_tokens(word, business) %>%
anti_join(stop_words, by = "word") %>%
count(biz_code, word) %>%
ungroup()

#pairwise_similarity of service words
closest_org_service_code <- grouped_service_code_words %>%
  pairwise_similarity(biz_code, word, n) %>%
  filter(similarity >= 0.5) %>%
  arrange(desc(similarity))

Build table to review business similarity (Check_closest_org_service_code). similar pair = 3887 (similar_pairs)

#To review pairwise_similarity in table format
Check_closest_org_service_code <- closest_org_service_code %>%
  left_join(grouped_biz_code %>% select(biz_code, business) %>% rename(business_1 = business), by = c("item1" = "biz_code")) %>%
  left_join(grouped_biz_code %>% select(biz_code, business) %>% rename(business_2 = business), by = c("item2" = "biz_code"))

similar_pairs <- Check_closest_org_service_code %>%
  filter(item1 < item2)

head(Check_closest_org_service_code, 100) %>%
  kbl(caption = "Preview of similar bussiness pair with similarity value more than 0.5") %>%
  kable_paper("hover", full_width = F) %>%
  scroll_box(width = "100%", height = "500px")

Regroup connected business

Simplified it to 102 Group of business. Group number 2 found to be densely connected whit 1169 business type, base one pairwise similarity of business description. Further classification required on Group 2.

Show the code

# Create an igraph object from your graph data
graph <- graph_from_data_frame(similar_pairs, directed = FALSE)


# Find the connected components in the graph
components <- components(graph)

# Get the membership of each node indicating its component
membership <- components$membership


# Count the number of nodes in each component
component_counts <- table(membership)

# Create a data frame with the component and count information
summary_table <- data.frame(Component = 1:length(component_counts),
                            Count = component_counts) 

summary_table <- summary_table  %>%
  rename(GroupNumber = Component, Count_business_type = Count.Freq) %>%
  select(GroupNumber,Count_business_type )%>%
  arrange(desc(Count_business_type))


# Print the summary table
head(summary_table, 200) %>%
  kbl(caption = "Preview of similar bussiness pair with similarity value more than 0.5") %>%
  kable_classic(full_width = F)  %>%
  scroll_box(width = "100%", height = "300px")

Preview of similar bussiness pair with similarity value more than 0.5
GroupNumber	Count_business_type
2	1169
14	9
25	8
24	6
10	5
11	5
17	5
22	5
40	5
53	5
27	4
32	4
39	4
82	4
1	3
3	3
16	3
20	3
23	3
28	3
31	3
36	3
37	3
42	3
44	3
45	3
51	3
56	3
66	3
72	3
74	3
75	3
84	3
90	3
92	3
93	3
4	2
5	2
6	2
7	2
8	2
9	2
12	2
13	2
15	2
18	2
19	2
21	2
26	2
29	2
30	2
33	2
34	2
35	2
38	2
41	2
43	2
46	2
47	2
48	2
49	2
50	2
52	2
54	2
55	2
57	2
58	2
59	2
60	2
61	2
62	2
63	2
64	2
65	2
67	2
68	2
69	2
70	2
71	2
73	2
76	2
77	2
78	2
79	2
80	2
81	2
83	2
85	2
86	2
87	2
88	2
89	2
91	2
94	2
95	2
96	2
97	2
98	2
99	2
100	2
101	2
102	2

Network Map without Group 2
Network Map of Group 2

Show the code

# Remove the nodes belonging to component 2
graph_filtered <- delete.vertices(graph, V(graph)[membership == 2])

# Plot 
plot(graph_filtered,vertex.size=8, vertex.label=NA, layout = layout_with_gem, vertex.color="lightsteelblue2", main="Network Map without Group 2")

Show the code

# Extract the subgraph for component 2
subgraph <- induced_subgraph(graph, V(graph)[membership == 2])

# Plot
plot(subgraph,vertex.size=8, vertex.label=NA, vertex.color="lightsteelblue2", main="Network Map of Group 2")

Community grouping

Show the code

ceb <- cluster_edge_betweenness(subgraph) 


plot(ceb, subgraph, vertex.size=8, vertex.label=NA, main="Exploring Community detection based on edge betweenness for Group 2")