<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Buddhima Naween Rathnayake on Medium]]></title>
        <description><![CDATA[Stories by Buddhima Naween Rathnayake on Medium]]></description>
        <link>https://medium.com/@buddhimanaweenrathnayake?source=rss-6be81e4f3487------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*MbPdgUVuhhEUrInn2Ed6lA.jpeg</url>
            <title>Stories by Buddhima Naween Rathnayake on Medium</title>
            <link>https://medium.com/@buddhimanaweenrathnayake?source=rss-6be81e4f3487------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sun, 22 Nov 2026 08:41:05 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@buddhimanaweenrathnayake/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Dialogflow — Capturing User Information and personalized responses]]></title>
            <link>https://medium.com/@buddhimanaweenrathnayake/dialogflow-capturing-user-information-and-personalized-responses-ed820a9d886d?source=rss-6be81e4f3487------2</link>
            <guid isPermaLink="false">https://medium.com/p/ed820a9d886d</guid>
            <category><![CDATA[chatbot-development]]></category>
            <category><![CDATA[dialogflow]]></category>
            <category><![CDATA[dialogflow-enterprise]]></category>
            <category><![CDATA[chatbots]]></category>
            <category><![CDATA[chatbot-design]]></category>
            <dc:creator><![CDATA[Buddhima Naween Rathnayake]]></dc:creator>
            <pubDate>Wed, 05 May 2021 02:28:42 GMT</pubDate>
            <atom:updated>2021-05-06T01:49:38.903Z</atom:updated>
            <content:encoded><![CDATA[<h3>Dialogflow — Capturing User Information and personalized responses</h3><p>Most of the time out chatbot should be capable of capturing user information and address user’s from their name in order to provide a personalized user experience. Dialogflow has the person Entity to overcome this, person entity can be added as a parameter to our training phrases in identifying that we are talking about a person.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Dv0NtN-lBZDTceaJV64-4A.png" /><figcaption>using sys.person entity to capture name of the user</figcaption></figure><p>Then you can use the person parameter to capture the user input and respond back to the chatbot response. use $ sign to return parameter values we specified through our entity.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/494/1*nXtuJSXwlzIB3FlsIENeEg.png" /></figure><p>Save your intent and “try it now”</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/350/1*-NbZY1RNCKzdJxVrT5fB8g.png" /></figure><p>This is simple and it works, however when it comes to Sinhalese names if we are only using an intent to capture the first name or last name Dialogflow won’t identify that they are related to the Person Entity. what you can do is use sys.any which will match to any non-empty input. however, it’s not a good practice to use sys.any without using fall-back intents.</p><p><strong>Capture both Last Name &amp; First Name scenario</strong></p><p>I’m using default welcome intent to show this, but you can replicate the senario in any intent you like.</p><p>to the welcome intent i’m going to add a follow-up intent as “Default Welcome Intent — First Name” this will capture the first name of the user. And i’m using sys.any to match with any input of the user.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*1AEdFLS-RXuQWuv69h_xQQ.png" /></figure><p>Then add another follow-up intent to the “Default Welcome Intent — First Name” as “Default Welcome Intent — Last Name” in order to capture the last name of the user. it will also have the same structure. but you have to change parameter names so you can differentiate them easily.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*-_k-MbhE1LCFDNRgvDCI8A.png" /></figure><p>As you can see in the output we have used a parameter in the former context and we have to specify it using a #</p><blockquote>#DefaultWelcomeIntent-Name-followup.person-firstname</blockquote><blockquote>$person-lastname</blockquote><p>“DefaultWelcomeIntent-Name-followup” is the follow-up context of the “DefaultWelcomeIntent-Name” Intent. so you need to specify the context where the parameter values are captured in order to display its value. however, we have to look at the lifespan of the output context</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/492/1*GLuGYPVQg0WKJsc80Kqxmw.png" /></figure><p>Once lifespan expires you won’t be able to access the parameter using the context. since at that moment, context is not available. say ‘hi’ to see our modified welcome intent in action. hope this helps :)</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=ed820a9d886d" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Using Events to trigger Dialogflow Intents — C#]]></title>
            <link>https://medium.com/@buddhimanaweenrathnayake/using-events-to-trigger-dialogflow-intents-c-4cac810268d5?source=rss-6be81e4f3487------2</link>
            <guid isPermaLink="false">https://medium.com/p/4cac810268d5</guid>
            <category><![CDATA[dialogflow]]></category>
            <category><![CDATA[chatbots]]></category>
            <category><![CDATA[chatbot-design]]></category>
            <category><![CDATA[chatbots-for-business]]></category>
            <category><![CDATA[chatbot-development]]></category>
            <dc:creator><![CDATA[Buddhima Naween Rathnayake]]></dc:creator>
            <pubDate>Wed, 06 Jan 2021 12:48:05 GMT</pubDate>
            <atom:updated>2021-01-06T12:48:05.797Z</atom:updated>
            <content:encoded><![CDATA[<h3>Using Events to trigger Dialogflow Intents — C#</h3><p>Events are used to trigger Intents without matching training phases. within the conversation, if you need to trigger a certain event without any user input you can use Events to do that.</p><p>If you check your Default Welcome Intent you can see it already configured with “Welcome” Event,</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/700/1*fOKgvD4FtmnO9zdbkKOSpA.png" /></figure><p>There are other platform-specific Events given to you, And you can also add your own custom events to trigger the desired intent as well.</p><p>Let’s look at a scenario where you need to redirect the user to the main menu at the end of the conversation. we need to create a new Intent with desired responses and add an event simply by typing the event name. I named mine as <em>Main_Menu.</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/717/1*Psh4KHBpkjKWWER8RbP6VA.png" /></figure><p>You can use the same code snippet we used to Detect Intent. However, this time we are sending <strong>GoogleCloudDialogflowV2beta1EventInput </strong>rather than sending a <strong>GoogleCloudDialogflowV2beta1TextInput</strong>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/555/1*1aDo5HYJhIRDAqjSBVwADA.png" /></figure><p>we don’t need to add any training responses to this intent since we are simply triggering it through “<em>Main_Menu</em>” event. However, if your Intent has Input Contexts you need to provide Contexts in order to trigger the intent. Let’s see how this can be achieved.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/706/1*3UvOy_kyIUYq0wBA-MuZkA.png" /></figure><p>I added <em>menu-input-context </em>to simulate the scenario. so we need to provide <em>menu-input-context </em>in order to trigger the “Go to Main Menu” Intent.</p><p>You need to add your input contexts as QueryParameters to the request.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/480/1*QCR05JyfvqowybMwzJuqQg.png" /></figure><p>“contexts” variable here is expecting a list of contexts, so we need to build the context list to be assigned to the “contexts”. The following method shows how to create a context list</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*jf_zmh38DuoB2xmANGHO4A.png" /></figure><p>I used a KeyValuePair to hold the context Name and the LifespanCount.</p><p>Key will be context name — <strong><em>menu-input-context</em></strong></p><p>and</p><p>Value will be the lifespan of the context</p><p><em>GenerateContexts </em>method will return a List of input contexts that are in “Go to Main Menu” Intent. you can trigger any event in Dialogflow using the above methods.</p><p>Good Luck :)</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=4cac810268d5" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Building your first chatbot using Dialogflow and C# — PART II (Implementation)]]></title>
            <link>https://medium.com/@buddhimanaweenrathnayake/building-your-first-chatbot-using-dialogflow-and-c-part-ii-implementation-bd696ba70462?source=rss-6be81e4f3487------2</link>
            <guid isPermaLink="false">https://medium.com/p/bd696ba70462</guid>
            <category><![CDATA[web-api-development]]></category>
            <category><![CDATA[c-sharp-programming]]></category>
            <category><![CDATA[dialogflow]]></category>
            <category><![CDATA[dialogflow-enterprise]]></category>
            <category><![CDATA[google-cloud-platform]]></category>
            <dc:creator><![CDATA[Buddhima Naween Rathnayake]]></dc:creator>
            <pubDate>Sat, 02 Jan 2021 17:17:23 GMT</pubDate>
            <atom:updated>2021-01-02T17:17:23.601Z</atom:updated>
            <content:encoded><![CDATA[<h3><a href="https://medium.com/@buddhimanaweenrathnayake/building-your-first-chatbot-using-dialogflow-and-c-63955570e03f?source=your_stories_page-------------------------------------">Building your first chatbot using Dialogflow and C#</a> — PART II (Implementation)</h3><p>In <a href="https://medium.com/@buddhimanaweenrathnayake/building-your-first-chatbot-using-dialogflow-and-c-63955570e03f">Part I</a> we covered basics of building a Dialogflow agent and how Intents play a major role in chatbot design. From this article we are going to implement intent detection scenario using ASP.NET web API.</p><p>This API acts as the middle layer between the website chat panel and the Dialogflow agent. when user types and input the message it will be sent to Dialogflow to detect the intent of the user phrase. we won’t be using sockets but eventually adding a message handler would be desirable. what we are going to cover here is simple intent detection request-response C# code.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/886/1*_NHJ5dowPzsfGVSrB-mUGw.png" /><figcaption>High level architecture of our chatbot</figcaption></figure><h3>Creating Chatbot API — ASP.NET Web API</h3><p>Open up Visual Studio and create a new project, choose ASP.NET Web Application in web section as follows,</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/940/1*Wa1xPBqihVBPU1Vf4tQ4WA.png" /><figcaption>create new project using VS</figcaption></figure><p>Give a name to your project. mine was ChatbotAPI and I’m going forward with .NET Framework 4.5</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/939/1*QpW-i-6StrpLmIty9DaTNg.png" /></figure><p>Click “OK” and choose the Web API project template, i didn’t choose any authentication since we’re not focusing authentication at this moment. later it would be better if you secure your API with a authentication mechanism.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/781/1*El4rWSNbgJMHqprhxLGC8Q.png" /><figcaption>choosing the Web API project template</figcaption></figure><p>Run your project and test out the implementation using Postman, Go to <em>ValuesController </em>inside <em>Controllers </em>using Solution Explorer. we are going to test the “Get Method” of our controller.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/335/1*peDTuNv4kBSAcVPgbgjkbw.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/424/1*fXQbkm6V_ApbS9kE0-LJrg.png" /></figure><p>So it should return string array containing “value1” and “value2”. open Postman and send a Get request to your values controller. and our API is working as intended.</p><p>Request URL : <a href="http://localhost:4145/api/values">http://localhost:4145/api/values</a> (change <a href="http://localhost:4145/api/values">4145</a> to your port number)</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/940/1*7U1Y80g3fQ-r8NhnwGyR-w.png" /><figcaption>postman GET request sample</figcaption></figure><p>Let’s Integrate Dialogflow API into our API, check their API documentation <a href="https://cloud.google.com/dialogflow/es/docs/reference/rest/v2-overview">here</a> we can use api V2 or V2Beta1, V2Beta1 is at Beta stage and it supports knowledgebases while api V2 doesn&#39;t have knowledgebase definitions.</p><p>Dialogflow service endpoint is at <a href="https://cloud.google.com/dialogflow/es/docs/reference/rest/v2-overview#service:-dialogflow.googleapis.com"><em>dialogflow.googleapis.com</em></a><em> .</em>we call this service on our own but it’s recommended to use client libraries provided by Google. since you don’t need to get your hands dirty :) get c# client library from <a href="https://cloud.google.com/apis/docs/cloud-client-libraries">here</a>, what I’m using is api v2beta1 since I need to dive into knowledgebases as well, you can get it from <a href="https://www.nuget.org/packages/Google.Apis.Dialogflow.v2beta1">nuget package</a> manager as well</p><blockquote>Install-Package Google.Apis.Dialogflow.v2beta1 -Version 1.40.2.1612</blockquote><figure><img alt="" src="https://cdn-images-1.medium.com/max/786/1*RcM1ZcZJDTc5B_8QFGv6kw.png" /><figcaption>package manager console output</figcaption></figure><p>And we are good to go..</p><p>we need to create a project in Google Cloud console since we need to authenticate our service through Google Cloud in order to use Dialogflow services. refer quick setup guide for more information <a href="https://cloud.google.com/dialogflow/es/docs/quick/setup">here</a></p><h3>Create Google Cloud project</h3><p>Go to Settings page on your Dialogflow console by clicking on settings ⚙️ icon. Check for Google Project ID</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/965/1*q2wbHFMjuFxLFg9Oq_Ar3g.png" /><figcaption>obtaining project ID</figcaption></figure><p>Go to Google Cloud Platform and explore your projects, Move to “ALL” tab if your Dialogflow project is not listed in Recent. It should be listed with your Project ID in Dialogflow console and click on the project. where you will be redirected to API Dashboard. Go to Credentials section and create a Service account.</p><p>Provide a service account name which describes the service. click on the “CREATE” button to proceed to create the service account and key.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/810/1*saK0ZjuLm9uscLlggZnV_A.png" /><figcaption>creating the Service account</figcaption></figure><p>and set the role to Dialogflow Admin using the dropdown.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/811/1*OPgnwCX4rq46PeBfnDlqgg.png" /><figcaption>granting service account access</figcaption></figure><p>Continue to grant service account access and skip Granting access to users or groups since this step is optional.</p><p>Open up the service account and scroll down to keys section and “Add KEY”</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/631/1*8bkoEGzn1804pwCnDnD2OQ.png" /><figcaption>creating API key</figcaption></figure><p>Select JSON key type and click “CREATE”</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/858/1*0OaArxq1Vt6Mz75fuLPPSA.png" /></figure><p>and your private key will be downloaded in JSON format. you can use this key by adding it to your solution.</p><h3>Detect Intent C# code</h3><p>I created a new controller in my solution as <em>IntentController</em> and it will be serving as the detect intent POST method. here’s the C# code used to create Google credential.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/581/1*TPGqmqOyVM4-KsDdGNepWw.png" /></figure><p>Here we should add reference to DialogflowV2Beta1 client library which we referenced earlier. Google Auth libraries are also added in this process as dependencies which will be facilitating the authentication. <em>configPath </em>is the path to the private key file we added to the project earlier.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/654/1*owrbJIwYfZoZ7wi6BgtVOg.png" /></figure><p><em>currentsession </em>is the session ID provided to Dialogflow at the beginning of the conversation. this is maintained throughout the conversation between Dialogflow agent and end-user. This session will expire in 30 minutes. Check Dialogflow session reference <a href="https://cloud.google.com/dialogflow/cx/docs/concept/session"><em>here</em></a>.</p><blockquote>A session ID is a string of at most 36 bytes in size</blockquote><p>So GUID is used as a session ID here.</p><p><em>projectId — </em>check private key JSON file created from Google Cloud Console and assign to <strong>project_id</strong> value to <em>projectId variable.</em></p><p>And we are done, Run your project and use postman to simulate the user input, here’s the response from <em>DetectIntent</em> method by providing “Hi” as Text value.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*3-dQvw7jW0V28QNJu6AhGA.png" /></figure><p><strong>QueryText</strong> — what we sent to the <em>DetectIntent</em> method (“Hi”)</p><p><strong>Action</strong> — action name of the Default Welcome Intent</p><p><strong>FulfillmentText</strong>—Response from Default Welcome Intent</p><p><strong>FulfillmentMessages</strong> — if we add multiple responses, array of responses will appear here.</p><p>Simply return <strong>FulfillmentText / FulfillmentMessages </strong>from your<strong> </strong>controller action and your chatbot is responding to end-user inputs now.</p><p>Finally enable Small Talk from Dialogflow console to have fun with your new chatbot :)</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/987/1*Lp__DJuOSBdBLEdd3hI6fA.png" /><figcaption>Enabling small talk</figcaption></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=bd696ba70462" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Building your first chatbot using Dialogflow and C#]]></title>
            <link>https://medium.com/@buddhimanaweenrathnayake/building-your-first-chatbot-using-dialogflow-and-c-63955570e03f?source=rss-6be81e4f3487------2</link>
            <guid isPermaLink="false">https://medium.com/p/63955570e03f</guid>
            <category><![CDATA[dialogflow]]></category>
            <category><![CDATA[chatbots]]></category>
            <category><![CDATA[dialogflow-enterprise]]></category>
            <category><![CDATA[c-sharp-programming]]></category>
            <category><![CDATA[chatbot-development]]></category>
            <dc:creator><![CDATA[Buddhima Naween Rathnayake]]></dc:creator>
            <pubDate>Fri, 01 Jan 2021 17:47:20 GMT</pubDate>
            <atom:updated>2021-01-06T01:54:11.298Z</atom:updated>
            <content:encoded><![CDATA[<p>Before we dig deep into the actual implementation let’s look at what Dialogflow is about. Dialogflow started off as Api.ai around 2014 which offered to build conversational interfaces. Google seeing the future bought the company around in 2016 and it&#39;s quite fast-growing.</p><p>Yeah..fast-growing and it seems Google is slowly introducing their monetization to the product. back in the day, you got a Standard version that was free forever, but now you get a Trial with a 300$ credit limit.</p><p>They also recently differentiated the product into CX (Customer Experience) and ES (Essentials). what they actually did was push the old intent-based architecture into ES edition and introduced CX edition which is suited for building more complex conversations using a state machine-based approach to the agent architecture.</p><p>And What we are going to talk about is the <strong>Dialogflow ES agent</strong>.</p><p>Let’s take a simple scenario where you need to book a flight. assume you talk to your travel agent and says “<strong>Hi, I need to purchase a ticket to Narita</strong>”, if you’re to replace the travel agent with a conversational bot. you need to understand the intent of the user first. that&#39;s the primary feature of Dialogflow that it provides us a way to map conversation into st of <strong>Intents</strong>. these intents are wrapped and packaged into one as <strong>Agent </strong>and the real world Entities like the <strong>Narita Airport </strong>will be mapped into <strong>Entities </strong>in Dialogflow.</p><p>Let’s see this in action</p><h3>Setting Up</h3><p>Go to <a href="https://dialogflow.cloud.google.com/">https://dialogflow.cloud.google.com/</a>, sign-in with Google which is obvious. you’ll be redirected into the Dialogflow console. and create your first agent.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/253/1*4DAr55j5f8_Bm5Y5piMDAg.png" /><figcaption>+ creating an agent</figcaption></figure><p>Create Agent with default settings and we’ll make the changes later once we need them. I created my agent as TravelAgent and you’ll be redirected to the Intents page. and you have Fallback Intent and Welcome Intent created by Default.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1007/1*P1YyOKE9ucQFIx-C48f4fw.png" /><figcaption>your intent defaults</figcaption></figure><p><strong>Default Fallback Intent — </strong>this is the intent that is called upon when the agent comes across a query/user input which is not mapped into any other Intents. this is already filled out with text responses such as “I didn’t get that. Can you say it again?”</p><p>read more on <a href="https://cloud.google.com/dialogflow/es/docs/intents-default#fallback">here </a>which will explain the negative example usage in fallback intents, which will be quite useful in your chatbot design.</p><p><strong>Default Welcome Intent —</strong> this is matched when the user starts the conversation, this intent is pre-populated with welcome responses such as “Hi”, “Hello” and this is also attached to welcome <strong>Event</strong>, we’ll talk about how to utilize events in future. let&#39;s keep it simple until then.</p><p>Let&#39;s try out these default intents in action,</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/370/1*3TlGyFy4tUbFLHR4BohKHg.png" /></figure><p>just type “Hi” in try out now, this acts as a simulator to try out your intents in the agent.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/337/1*pImAk3FdrmF5Nd44qFbhyw.png" /></figure><p>Default response is the pre-populated response from Dialogflow. If you added a response from a different channel (eg: Google Assistant) simply click on the dropdown and select the respective channel. then the response from that channel will be shown. as you can see simulator provides the Intent which was called upon and the Action. Diagnostic info is for checking how the JSON response from Dialogflow would turn out.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/781/1*8w1tAyYQXaLB1FRCNYC5xw.png" /></figure><p>At this point you’ve basically created your bot, what you need to do next is simply building up your Intents to match the requirements of the chatbot as to what end-user inputs the chatbot should provide a response to.</p><p>Click on your <strong>Default Welcome Intent </strong>and open it up, we are going to see how an intent works, First off you have Contexts and Events. <strong>Contexts</strong> are very important aspects of any conversation. if we go back to our travel agent example (“<strong>I need to purchase a ticket to Narita</strong>”) the context here would be purchasing a plane ticket. however, if the travel agent asks for your personal details as “<strong>Please tell me your Name</strong>” the context of the conversation change to get personal details. these input and output contexts act as a bridge between the Intents. and the context would change once you get from one Intent to another.</p><p>we’ll discuss this more later while we build up the chatbot. let&#39;s focus on <strong>Training phrases</strong> and <strong>Responses</strong> for the moment. Training phrases are simply the users&#39; inputs that you can expect the user will type into the chatbot and which will trigger the respective intent. in <strong>Default Welcome Intent </strong>“Hi” is a Training phrase, because it triggers the welcoming intent of our agent. you can see pre-populated Training phrases from Dailogflow in the welcome intent.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/642/1*atfKtA6URSyoRpIWRACO9A.png" /></figure><p>Moving to the Responses section, it’s simple what chatbot will respond to the end-user input.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/661/1*O-Ykj8Zfq-UFQpXmTNYWgQ.png" /></figure><p>It’s basically a request-response scenario where the request/Training phrase (end-user Input) would be “Hi” and the response to the welcome would be “Hi! How are you doing ?” you can add multiple responses to an Intent as shown here.</p><p>What you should put your focus into is the conversation flow of your chatbot. in the welcome scenario you welcome the user as “Hi! How are you doing ?” and the next response from the user would be an answer to the question “How are you doing ?” it’s best to go with responses like “How can I assist?” since it’s more aligned with the purpose of our chatbot — assisting the user to complete his/her task.</p><p>we’ll cover chatbot design and implementation of C# API to detect intent in the next article. hope you guys now have an idea about Dialogflow. which will be crucial in moving to the next section of our article series.</p><p>Have a look at Part II — <a href="https://medium.com/@buddhimanaweenrathnayake/building-your-first-chatbot-using-dialogflow-and-c-part-ii-implementation-bd696ba70462">Implementation</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=63955570e03f" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>