/hackathon-product

Primary LanguageJavaScript

Advendetour

Grand-prize winner for AngelHack 2016 - Montreal

Team members: Jing Liu, Jordan Quan, Paige Sun, Robin Luo, Yuning Bie

Workflow Introduction & Introduce Pain Point -------------------------------------------------------- Have a Point A and a Point B? Well, Advendetour is here to show you that a straight line is NOT the best way to go. Advendetour will suggest to you places to visit and you can have your journey, your way.

Use Case Example: Billybob is travelling in Europe, arriving in Paris and flying out of Rome in 3 weeks. But Billybob doesn’t want to just go straight from France to Italy, Billybob wants to visit the sights along the way! But, Billybob doesn’t know where to go!! Advendetour will suggest places to stay and things to do along the way. (*demo while presenting situation and function of program: explain how the cities are chosen, activities and hotel prices are classified)

Catchphrases: Spend less time planning, and more time travelling. The road goes ever, ever on.

Architecture ----------------------------------------------- Our project takes in the starting city and an ending city, and iteratively return all the spots of attraction, including details such as price and “things to do”, within certain range, and meanwhile plans a route and shows directly on google map. Our algorithm constantly update our suggested path and attraction spots according to previous user choices.

How our algo works: We initialize the graph with the start/end point input by user, and a variance number that represents how further away the user is willing to travel that would deviate from his/her final goal. Then we take points on the minimum route, using Expedia API to sort out all attraction places/cities according to how far away from the min route, and present the best 5 choices to user. After the user chooses his favourite, that choice become the current node upon which we update our graph. We run the same algo iteratively until the user is close enough to the destination point, or the user wants to finalize his trip.

Process: User inputs start and destination and how far they think they can travel in a day. We then determine the likeliest stops for such a journey, suggesting cities to stay at and fun things to do! They can view their growing itinerary on googlemaps and in a printable format online.

Selling Points & Future Implementation -----------------------------------------------------

Intuitively build an adventure trip within minutes! Get to your destination on time and within budget Experience local culture Would like for the user to edit their route, if they particularily want to visit a specific city. Wanted to keep a running tally of expenses and trip duration (both for hotel stays and activities) Wanted to provide alternative options for hotels and prices, and even ways to book online!

Conclusion/Future improvement Add time calculator feature that will make sure that the user do not exceed his/her time limit that he/she sets, make sure that he/she does not choose too many activities Show more details/choices about activities and hotel prices (be able to choose number of rooms, date the user arrive there)

@ helper functions

func distance(start-lat, start-lon, end-lat, end-lon) -> distance Returns distance between two points

coord(cityName) → [lat, long] returns latitude and longitude of start-city and end-city, which is passed as parameters for queryForAllCities() Sample: start-lat, start-lon = 48.852739, 2.345838 end-lat, end-lon = 41.889698, 12.476393

difflatlong(city1,city2) -> [diff in lat, diff in long] totaldiff(city1,city2) -> distance between city1 and city2 in KM

  • getCheapestHotel( [hotelModels] ) -> hotelModel

@ global variables

@ Frontend: --------------------------- Web Browser View -------------------------------- Start: _______________________ Destination: __________________ Number of Days: ______________

User clicks “Let’s Go!” and ends up in page 2

Upon clicking on “Let’s Go!” the frontend sends to backend a userQueryModel userQueryModel -------- done once at the beginning Start-city = Paris End-city = Rome distancePerDay = 400km ←-- make sure to type in a much bigger number than 100km How much do you want to drive in a day? ( in km)

@ Backend

Gets userQueryModel from frontend //DONE

func queryForCities(userQueryModel) → [5CityModels] Var breakPoint[latitude, longitude]

queryCitiesInBetween(start-city, end-city) → JSONObjects getBreakpointBetweenTwoCities(start-lat, start-lon, end-lat, end-lon) → breakPoint Sample: getMidpointBetweenTwoCities(48.852739, 2.345838,41.889698, 12.476393) → (25.5992885,27.183045500000002) var distanceFromMidpointToEnd = getDistance(breakPoint, end-lat, end-lon) var radius = min(distanceFromMidPointToEnd, 100km) query for all cities within a radius from center Use Find geographical features with a radial distance from a point defined by the lat and lng parameters Within = radius type = “city” Lat & lon = mid-lat & mid-lon Sample: http://host:port/features?type=city&lat=25.5992885&lng=27.183045500000002&within=14.7066525km&apikey=NH6oq54KA957fmLpv3x1pFx8CE1xhRoc

getCitiesInBetween(JSONObjects) -> [cityModels] //DONE Return [cityModels] One CityModel Object: name latitude longitude distance = 0 calcDistanceForCities([cityModels], breakPoint) //DONE For every city in [cityModels] var city-lat = city.latitude var city-lon = city.longitude city.distance = getDistance(breakPoint, city-lat, city-lon) sortAllCities ([cityModels]) → [5CityModels] //don’t need mid-points as argument because distance is in field distance //DONE Sort all the cityModels by their distance to the midpoint ascending order and return the first [5CityModels] /*get5CityModels(mid-lat, mid-lon,[cityModels]) -> [5CityModels] */

getHotelsForCities([5CityModels]) For city in [5CityModels] city.cheapestHotel = getCheapestHotel(cityName) hotelModel: Name, price

@ Frontend --- displays [5CityModelsForDisplay] in rows that automatically pops up [5CityModelsForDisplay] An array containing 0-5 cities Tap a city to find more cities around that city On each row, display: City-name ---------------- [button to search for activities]

Page 2: ---------------------------- Web Browser View ------------------------------------------------ start-city between-city 1 between-city 2 between-city 3 between-city 4 between-city 5 end-city

--- Sent from frontend to backend:

userSelection selected-activity --- null if the user selected a city selected-city --- null if the user selected an activity

Sent from backend to frontend [JSONObject]

Convert from [JSONObject] to [nodeObject]

nodeObject

activityModel

APIs

Activity search: Pros: can get location, description, and price Cons: can only search using http://terminal2.expedia.com:80/x/activities/search?location=Montreal&startDate=2016-06-04&endDate=2016-06-04&apikey=pvTBSyAhBBWKLgnGxW2h8LzOIjqEGoGM

Extra info: Paige’s API key: pvTBSyAhBBWKLgnGxW2h8LzOIjqEGoGM Jing’s API key: NH6oq54KA957fmLpv3x1pFx8CE1xhRoc

GoogleMap API key: AIzaSyB5LyIEhdrka55mYbw5pIHSwDyA-FR82gw

API CALLS (as experimented with by Jordan)

Geography Search for cities, based on a lat/long coordinate, and search radius, looking for cities in that region http://terminal2.expedia.com:80/x/geo/features?within=30km&lat=37.7874&lng=-122.4082&type=city&apikey=API