Tag Archives: watson

Using IBM Watson Language Translation Services with Python

Here is a very simple example using Python of calling the IBM Watson Language Translation Service. Feel free to use and modify the code as needed. You will need an IBM Bluemix account in order to use the translation service.

# CTEC 121 Intro to Programming and Problem Solving
# Bruce Elgort / Clark College
# Using IBM Watson's Language Translator
# February 27, 2016
# Revised: May 24, 2016
# Version 1.1

import requests

def cls():
    print("\n" * 5)

def translate_text(text,source,target):
    username = 'your username'
    password = 'your password'
    watsonUrl = 'https://gateway.watsonplatform.net/language-translation/api/v2/translate?source=' + source + '&amp;amp;target=' + target + '&amp;amp;text=' + text
    try:
        r = requests.get(watsonUrl,auth=(username,password))
        #print(r)
        return r.text
    except:
        return False

def welcome():
    message = "Welcome to the IBM Watson Translator\n"
    print(message + "-" * len(message) + "\n")
    print("Have fun!\n")

def main():
    cls()
    welcome()

    data = input("Enter some text to be translated:\n")

    print()
    print("What language should I translate it to?")
    print("1) Spanish")
    print("2) Arabic")
    print("3) French")
    print("4) Portuguese")
    print()
    target = input("Select a language from the list above: ")

    if target == "1":
        target = 'es'
    elif target == "2":
        target = 'ar'
    elif target == "3":
        target = 'fr'
    elif target == "4":
        target = 'pt'

    results = translate_text(data,'en',target)
    print()
    print("Here is the text translated for you:")
    print(results)

main()

Using IBM Watson Tone Analyzer with Python

Here is a sample Python program that you can use to analyze the tone of text using IBM’s Watson Tone Analysis service on Bluemix. To get started with Bluemix you can sign up at http://bluemix.net. Feel free to use and modify the code as needed.

# CTEC 121 Intro to Programming and Problem Solving
# Bruce Elgort / Clark College
# Using IBM Watson's Tone Analyzer to detect and interpret emotional, social, and writing cues found in text.
# February 26, 2016
# Version 1.0

import requests
import json

def analyze_tone(text):
    username = 'your username'
    password = 'your password'
    watsonUrl = 'https://gateway.watsonplatform.net/tone-analyzer-beta/api/v3/tone?version=2016-05-18'
    headers = {"content-type": "text/plain"}
    data = text
    try:
        r = requests.post(watsonUrl, auth=(username,password),headers = headers,
         data=data)
        return r.text
    except:
        return False

def welcome():
    message = "Welcome to the IBM Watson Tone Analyzer\n"
    print(message + "-" * len(message) + "\n")
    message = "How it works"
    print(message)
    message = "Perhaps a bit too aggressive in your emails? Are your blog posts a little too friendly? Tone Analyzer might be able to help. The service uses linguistic analysis to detect and interpret emotional, social, and writing cues found in text."
    print(message)
    print()
    print("Have fun!\n")

def display_results(data):
    data = json.loads(str(data))
    print(data)
    for i in data['document_tone']['tone_categories']:
        print(i['category_name'])
        print("-" * len(i['category_name']))
        for j in i['tones']:
            print(j['tone_name'].ljust(20),(str(round(j['score'] * 100,1)) + "%").rjust(10))
        print()
    print()

def main():
    welcome()
    
    data = input("Enter some text to be analyzed for tone analysis by IBM Watson (Q to quit):\n")
    if len(data) >= 1:
        if data == 'q'.lower():
            exit
        results = analyze_tone(data)
        if results != False:
            display_results(results)
            exit
        else:
            print("Something went wrong")

main()

Slack Bot – Slack Channel Tone Analysis Using IBM Watson

In this video, I demonstrate how I used IBM Watson’s Tone Analyzer service to analyze an entire Slack channels’ textual content history. The bot allows you to select the channel you want to have analyzed and then presents charts displaying the various tones in the channel.