While using Knowledge Base in Dialogflow, $Knowledge.Answer[1] is returning a response whereas $Knowledge.Answer[2] or $Knowledge.Answer[3] are not working. Any idea?
According to googles Dialogflow documentation.
You can see from the documentation that , Multiple responses are triggered only when you have multiple answers for the same question. And how do you give multiple responses?
Well this bugged me a lot too. But it is simple. Just give the same question twice with different answer or give a slightly different question with different answer.
Example(FAQ.txt / UTF-8):
How big is google?,Google is the universe.
How big is google?, It is the biggest in the world.
Now go to Dialogflow console and type in this question. And tada. You get two responses.
This is a sample response which I am getting for the question I asked.
So Knowledge.Answer[1] and Knowledge.Answer[2] and so on, is triggered only when there are multiple answers to that question.
Hope this helps.
There's a Bike Shop sample on Github that gives a walkthrough of how to get knowledge connectors setup for your agent via uploading a .csv file as the data source.
Related
With the wake of the pandemic causing schools to go to distance learning, many classes take attendance by using a simple google form sent out to students to complete for each class everyday. While this seems like a simple solution, it is a pain for students to complete and keep track of. One way that I thought I could make this easier would be to keep track of which forms I have submitted everyday.
As of now, my problem is that I need a way to subscribe to the submit of a google form (based on a link). When that google form is submitted all I need to do is find a way to convey that to a program. What I do not understand is how I would be able to do that without having ownership of the form or make a teacher recreate the form. Is there a way that I can check if a google form has been submitted?
A couple of ideas I have had would be to sniff network traffic for a post request from a google form and get that link and compare it to other links in the program to see which one was submitted, but I would think there is an easier way to do this. Any ideas or code is welcomed.
I understand stack overflow is for already written code so if you do not agree with this post either ignore it or point me to the correct place where this should be posted. Thank you.
I am making a chat bot to answer questions on a particular subject(example, physics). How would you structure all the possible questions as intent in dialogflow?
I am considering the following 2 methods,
Methods:
make each question as an unique intent.
group all the questions into one "asking questions" intent and use entity to identify the specific question being asked.
Pros:
Dialogflow can easily match users input to the specific questions using low confidence score threshold, and can give multiple training phrases per question.
Only need one "asking questions" intent, neater and maintaining it is easier.
Cons:
There will be tons of intents, and maintaining it might be a nightmare. Might also reach the max number of intents.
Detecting entity might be more strict and less robust.
I would suggest you to try Knowledge Base feature of DialogFlow.
You can give multiple web-page links from where it can gather all the questions, or you can manually prepare a list and upload it to DialogFlow.
That way you don't need to make it in separate intents, it will try to match it automatically.
Let me know if you have any confusion.
This looks like an FAQ type chatbot. You can develop the chatbot in 2 ways:
Use Prebuilt Agents - Go to prebuilt agent and select and import FAQ and add your intents.
Use Knowledge Base approach - This is in Beta mode right now, but super easy to build.
a. You need to enable Beta Features from the agent settings
b. Go to Knowledge Base on the left menu, create a new document and upload CSV file (Q and A). You can also provide a link for Q/A if you have.
Check out the documentation for more details.
Knowledge Base seems to be the best way, but it only supports English content
I've proposed a title for our thesis, Movie Success Prediction through Social Media comments using Sentiment Analysis, is there a way you can get the comments on social media (twitter, Instagram, Facebook etc.) and use it for your software? like an API or any other way. is that even possible to use your software on different social media to get the comments for prediction or should i change my title and stick to one social media like Facebook or twitter only?
what's the good algorithm for this?
what programming language and framework/IDE should i use?
I've done lots of research on google and still hoping for more info here. Thank you.
Edit: I'll only use YouTube and YouTube API.
From the title of your question, it seems that the method you need to use is distant supervision. You need to retrieve data with labels you think it is proper for your task. For instance, a tweet containing #perfect hashtag would probably be a positive tweet. So, you can define set of hashtags for your task, negative, positive or even for neutral; then you can retrieve tweets by those via Twitter API. For your task, those should be for movies, therefore your data should contain movie related information in first place.
Given that you will deal with text data and you'd like to create your own dataset, it is better to start with Twitter. Its API works for your needs and it is very well-documented. The language and frameworks are upto your choice, since APIs supports many known languages as well. Personally, I'd start with python or java to quickly solve future problems easier with community support.
For a general survey of this area, you may dive into papers and resources from here:
https://scholar.google.com.tr/scholar?hl=en&q=distant+supervision+sentiment+analysis
Distant supervision could be used to create a sentiment lexicon out of millions English tweets by using sets of negative and positive hashtags as well. You may take a look at Chapter 5 of this thesis ( https://spectrum.library.concordia.ca/980377/1/Ozdemir_MCompSc_F2015.pdf ), this may also give a good insight for your thesis, too.
Hope this helps.
Cheers
I'm using wit to recognize different intents in a retail context. Some of them trigger (successfully) FAQ answers, other initiate a business logic.
Surprisingly, I'm having a lot of trouble with the most basics conversational intents, like answering a hi or hello. Specially if they come as a single word (it doesn't get hi or hello but it successfully returns the correct intent for hi buddy or hey dude). Obviously there's a high chance that the first thing an user would say is just a simple hello, any of you found the same issue? Any guidance on that?
It is actually the first time I experience this issue, and I haven't heard about it. Could it be related to the increasing number of intents created (now 15+)? I'm using trait as a search strategy.
Greetings intent
Click on image for a larger version of the image.
Thank you very much for your help,
I am trying to build a related posts functionality like what we see here in stackoverflow on the bottom right corner. My main difficulty is that the related posts have to be determined at runtime when loading a particular post.
I am thinking to look for posts with similar or equal set of tags, also similar titles and possibly similar sets of keywords in the main post content, the problem is that the more signals you look for the slower the database will be to return data.
I have also thought about using Google Site Search (we have an account) as Google does understand the relationship between posts very well, but unfortunately the related: parameter is broken according to Google.
I am looking for ideas on how to best achieve this. Has any of you ever done something like this? What's the best way to achieve this type of thing?