Appointment Booking Bot Assistant in Mobile App

Appointment Booking Assistant Bot in Mobile App

Booking appointments via chat competes with native calendar pickers. To win, the dialog must be substantially shorter: the user describes desired time in words, the bot suggests suitable slots.

Parsing Time Expressions

The main technical challenge is understanding arbitrary time expressions: "tomorrow after noon", "Friday morning", "next week, preferably evening". This is an NLP task.

For Russian, the Natasha Python library extracts dates and times well from unstructured text:

from natasha import Segmenter, MorphVocab, NewsEmbedding, NewsDatesExtractor

segmenter = Segmenter()
morph_vocab = MorphVocab()
emb = NewsEmbedding()
dates_extractor = NewsDatesExtractor(morph_vocab)

text = "book me next Friday at 3pm"
for match in dates_extractor(text):
    print(match.fact)  # DateFact(year=2024, month=1, day=19, hour=15, minute=0)

With Dialogflow — built-in system entities @sys.date, @sys.time, @sys.date-time work for Russian. For Rasa — duckling extractor running as a separate HTTP service.

After parsing: request available slots from the scheduling system, offer nearest to requested time.

Scheduling System Integration

The bot works via the scheduling system API. Popular options in CIS: 1C:Enterprise (healthcare, services), YCLIENTS (beauty), Calendly API, Google Calendar API, custom systems.

For Google Calendar:

const { google } = require('googleapis');
const calendar = google.calendar({ version: 'v3', auth });

async function getAvailableSlots(serviceId, date) {
  const freebusy = await calendar.freebusy.query({
    requestBody: {
      timeMin: dayjs(date).startOf('day').toISOString(),
      timeMax: dayjs(date).endOf('day').toISOString(),
      items: [{ id: serviceCalendarId }]
    }
  });
  // Calculate free slots between busy times
  const busySlots = freebusy.data.calendars[serviceCalendarId].busy;
  return computeFreeSlots(busySlots, workingHours, serviceDuration);
}

Critical: handle the race condition where a slot is taken between the bot's offer and user confirmation. Solution — two-phase booking: temporary reservation for 3–5 minutes when showing slot, confirmation on agreement.

Bot UI for Appointments

Text dialog is supplemented with inline components:

Horizontal date scroll — when user hasn't specified a date, show next 7 days as chips. Tapping a date requests slots for that day.

Time grid — display available slots as buttons in a grid; occupied ones are inactive. On Android, use FlexboxLayout with dynamically added button chips. On iOS — UICollectionView with compositional layout.

Confirmation card — specialist, service, date/time, duration. "Confirm" and "Change" buttons.

Reminders

After successful booking, the bot schedules a reminder. Options:

• Push notifications 24 hours and 2 hours before — via FCM/APNs
• SMS via Twilio or SMS.ru — for those who disabled push
• Add to device calendar — EventKit on iOS, CalendarContract on Android

Users choose method at confirmation.

Development Process

Analyzing target scheduling system, API documentation.

Developing NLP logic for time parsing and slot selection.

Server side: dialog state machine, calendar/booking API integration, slot reservation.

Mobile UI: inline date/time selection components, confirmation card.

Testing with real edge cases: all slots occupied, working/non-working days, midnight transitions.

Timeline Estimates

Appointment bot with basic NLP and Google Calendar API plus mobile UI — 1–2 weeks. With custom scheduling system, multiple specialists/services, SMS reminders — 3–4 weeks.