Closed. This question needs to be more focused. It is not currently accepting answers.
Want to improve this question? Update the question so it focuses on one problem only by editing this post.
Closed 4 years ago.
Improve this question
We have a number of chatbot instances we are testing on Dialogflow, some of which are developmental ones and others which are locked down production instances. We do extensive testing to make sure our responses are consistent and correct on a frequent basis.
We've noticed that even on a locked down instance where we aren't changing anything, the intent and entity responses from Dialogflow on an instance can change over time, and in some cases become incorrect.
This suggests that the underlying training algorithm is changing, and that instances are being auto-trained when such changes are released. Does anyone know if this is the case? If so, are there any suggestions about how to maintain a stable instance?
I am in the same situation. We have turned ML off and auto expansion off because when we go into the entities we sometimes get test data being added in there.
First thing, your intent & entity responses will never change if you have locked down instances unless & until you're asking a valid/relevant question. You keep it as it is for as much time as you want, your instances won't get affected.
Second thing, you have a fear of instances are getting auto-trained. See, there is basic difference between AI & ML. Replying user by understanding the context of what user wants to say is AI, on the contrary, ML comes in picture when you're trying to learn from user says & answer based on that. In api.ai, NLP is using AI & not ML so there is no question of auto-training. Now enable/disable option for ML in api.ai is only for computing threshold of how much percent user entered query matches to user says that you have in intent & not for auto-training.
Related
Closed. This question needs to be more focused. It is not currently accepting answers.
Want to improve this question? Update the question so it focuses on one problem only by editing this post.
Closed 5 years ago.
Improve this question
What are some ways including machine learning that I can use in my projects to generate things related to another. Like related apps, related websites, related products, etc.
I've been brainstorming these are strategies...
one way i can think of is show items from same category. But that would be too broad.
2nd way improves upon previous step, it's to keep track of what people click next and promote that item. Meanwhile keep bottom list randomized to let other relevant items show up and get clicked.
3rd way is to use machine learning and provide training data somehow and use that.
I want something simple but smart, as it gets better with time.
Collaborative filtering is designed for solving exactly this problem. The problem with this approach is that produces good results having a lot of data only. I mean... A LOT. And it's not a really simple thing to use. However, any machine learning technique is not simple. There are some node.js packages for CF available, but I have no idea how good are they.
Closed. This question needs to be more focused. It is not currently accepting answers.
Want to improve this question? Update the question so it focuses on one problem only by editing this post.
Closed 6 years ago.
Improve this question
I'm pretty new on search engines and pretty newbie on machine learning. But I wanted to know if there is a way to combine functionalities of search engines like elasticsearch or Apache Solr and machine learning project like Apache Mahout, H2O or PredictionIO.
For exemple, if you work on a travel website where you can search for a destination. You start type "au", so the first suggestions are "AUstria", "AUstralia", "mAUrice island", "mAUritania"... etc... This is typically what elasticsearch can do.
But you know that this user has already travelled on Mauritania three times, so you want that Mauritania goes on the first place of suggestions. And I guess that's typically what machine learning can do.
Is there bridges between this two type of technologies ? Can machine learning ensure the work of search engine efficiently ?
I'm open to all answers, regardless of the technologies used. If you have ever experienced this type of problems, my ears are wide open :-)
Thank you
Your question is very general in nature- so my answer will have to be the same.
Consider a recommender framework such as the one in Apache Mahout correlated co-occurance. Unlike the vanilla spark recommender, this implementation allows for multiple types of actions, such as viewed a web site, booked a trip their before, demographic information, etc.
Now you would calculate the recommendations for each user at whatever interval. Recommendations being based on multiple criteria and what other people similar to this user has done. Consider your 'items' in this case to be every destination in the world. So we now have every possible destination ranked for each user.
It is then a trivial extension to index elastic search by user/the ordered list of that users recommended destinations.
For example, we have a user who has visited Berlin, looked at several hotels in Vienna, and is from Romainia. When the user types in "au", we would expect to see "Austria" come up in the results much higher than 'Austrailia'
Per the comments and down votes- you probably should have either A) asked a more specific programming question or B) asked this question on another forum such as Data Science Stack Exchange, fyi
Closed. This question is opinion-based. It is not currently accepting answers.
Want to improve this question? Update the question so it can be answered with facts and citations by editing this post.
Closed 5 years ago.
Improve this question
This may be a weird question and please bear with me, I am completely new into this.
I have a list of 20 requirements (use cases) which I received from my client. With him, I prioritized this list of requirements (1 highest 3 lowest). I wrote for every requirement a use case scenario (rather than user story). I also have a use case diagram and some technical designs (class diagram, database diagram).
Now, my plan is to separate this list of 20 requirements into 5 sprints. Each sprint lasts one week.
During every meeting with my client, I can show the product with 4 new use cases implemented. If one of them isn't finished, I move it to the next sprint and my client can request a change during this meeting. During this change, the specific use case diagram and classdiagram/database diagram may be changed.
Is this considered to be Agile? (Even though he gave me the full 20 requirements from the start of the project)
Agile is sort of a big tent, but I would not apply that label to the process you've described. You are describing lots of upfront design work and a full specs up front. The schedule assumes all the req take the same amount of time to implement, thought you acknowledge that it could slip.
The primary agile feature I see is the tight (weekly) feedback loop with client.
I recommend trying on http://pm.stackexchange.com.
This is not considered as Scrum:
- Schedule is prefixed (5 sprints).
- Velocity is prefixed (4 use cases/sprint).
- No scrum ceremony is followed as such.
- All requirements are given upfront.
Please refer - https://www.scrumalliance.org/why-scrum/core-scrum-values-roles
Closed. This question needs to be more focused. It is not currently accepting answers.
Want to improve this question? Update the question so it focuses on one problem only by editing this post.
Closed 9 years ago.
Improve this question
We are building a text search solution and want a way to measure precision and recall of the system every time we add new document types. From reading some of the posts here it sounds like a machine learning based solution is the way to go. Can a expert comment on this? We will then look to add machine learning folks to our team.
The only way to get the F1-score require knowledge about the correct class, rank of all samples obtains by evaluation querys, and you also need thoses evaluation querys.
Any machine learning will need a large quantity of manual work to provided thoses samples and/or querys. So large that it wont save you any time.
Another bad aspect of this evaluation is through to learning-related intrinsic errors. It will go with the growing size of the index of the search engine and the number of examples required. You never get a good evaluation.
Forget machine-learning for the evaluation of search engine.
Build by hand your tests querys and sample, by the time it will become big and reliable.
If you really want machine-learning in your system, you should look at query pre-processing. Getting some meta-information about the query by another way (you say SVN, why not?) is generaly a good for performance and while it did'nt change the result, you can use the same sample for an end-to-end evaluation.
That what I have done few years ago, but with naive baye classifier on natural langage analysis.
Closed. This question is opinion-based. It is not currently accepting answers.
Want to improve this question? Update the question so it can be answered with facts and citations by editing this post.
Closed 8 years ago.
Improve this question
My current work involves working on a large number of bugs.
We normally (non TFS) would add these to an iteration backlog (grouped into Stories) after estimating and prioritising; then work through, mark actual effort.
I want to try and understand how I would work on these bugs using the TFS Agile template as intended. But am really struggling to find best practices and examples specifically for bugs for the TFS Agile template in TFS2010.
Cheers, Nick
I hear some parts in your question:
"Add these to an iteration backlog": you can use the iteration path of the work items. Best practice is to create an iteration called backlog.
"Grouped into Stories": In TFS 2010, the default traceablity is that on a User Story you define the Test Cases which validate the User Story. The Bugs are reported against the Test Cases.
"Estimating": You can use the Remaining work field for that
"Prioritsing": You can use the Stack Rank field
"Mark actual effort": Use the Completed work field
What we have been doing is:
Raising bug during testing by a tester.
During iteration planning we may decide to allocate X amount of time to fix outstanding bugs, so we creat a bug fixing story for that iteration of X story points.
Bugs are chosen that we think should be fixed within the iteration, a task is created for each bug along with a time estimate and any high level technical details. Note the task is created as a child of the story and also related to the bug.
The key is that bug work items are not developed against directly, a related task is.