How to put the text array into one text ? i wanna deal with the final aggregate text with llm.
You should be able to use the Array Aggregator on that since its a bundle.
This is incorrect advice for two reasons:
-
it still doesn’t combine the two bundles into a single text/string variable
-
the module is already an aggregator
You are already using a Text Aggregator, which should combine the bundles if you have set the source module correctly to where the bundles are being generated from (which is NOT the LLM module btw).
For further assistance, please provide the following:
- Screenshot of the ENTIRE scenario
samliew – request private consultation
Join the Make unofficial Discord server!
1 Like
Thanks for the information, but you have not provide any of the requested information yet 
It is, because what is being aggregated?
You might need a second text aggregator, but I can’t see your full scenario.
If you would like assistance, please provide the following:
1. Screenshots of module fields and filters
Please share screenshots of relevant module fields and filters in question? It would really help other community members to see what you’re looking at.
You can upload images here using the Upload icon in the text editor:
2. Scenario blueprint
Please export the scenario blueprint file to allow others to view the mappings and settings. At the bottom of the scenario editor, you can click on the three dots to find the Export Blueprint menu item.
(Note: Exporting your scenario will not include private information or keys to your connections)
Uploading it here will look like this:
blueprint.json (12.3 KB)
3. And most importantly, Input/Output bundles
Please provide the input and output bundles of the modules by running the scenario (or get from the scenario History tab), then click the white speech bubble on the top-right of each module and select “Download input/output bundles”.
A.
Save each bundle contents in your text editor as a bundle.txt file, and upload it here into this discussion thread.
Uploading them here will look like this:
module-1-input-bundle.txt (12.3 KB)
module-1-output-bundle.txt (12.3 KB)
B.
If you are unable to upload files on this forum, alternatively you can paste the formatted bundles in this manner:
-
Either add three backticks
```before and after the code, like this:```
input/output bundle content goes here
``` -
Or use the format code button in the editor:
Providing the input/output bundles will allow others to replicate what is going on in the scenario even if they do not use the external service.
Following these steps will allow others to assist you here. Thanks!
samliew – request private consultation
Join the Make unofficial Discord server!
1 Like
my Scenario blueprint will contain apikey ，so itis not confinet todothat
output：
[
{
“text”: “As we approach the year 2024, the topic of Semantic Search and Natural Language Processing (NLP) in AI Search Engines is gaining significant attention and discussion. With the increasing demand for more intelligent and intuitive search experiences, semantic search and NLP technologies have become essential components of AI search engines. These advanced technologies enable AI search engines to understand the context and intent behind user queries, providing more accurate and relevant results.\n\nMany experts and industry leaders have highlighted the importance of semantic search and NLP in AI search engines. According to a report by MarketsandMarkets, the global AI in search market is expected to grow from $1.14 billion in 2018 to $4.52 billion by 2024, at a compound annual growth rate (CAGR) of 28.5% during the forecast period. This growth is primarily driven by the increasing adoption of AI-powered search solutions in various industries, including e-commerce, healthcare, and finance.\n\nSemantic search and NLP technologies are also becoming increasingly important in the field of conversational AI. With the rise of voice assistants such as Amazon Alexa, Google Assistant, and Apple Siri, there is a growing need for AI search engines that can understand natural language queries and provide accurate and relevant results. According to a report by Grand View Research, the global conversational AI market size was valued at $4.2 billion in 2019 and is expected to grow at a CAGR of 21.9% from 2020 to 2027.\n\nFurthermore, the integration of semantic search and NLP technologies in AI search engines can also help to improve the user experience by providing more personalized and relevant results. By understanding the context and intent behind user queries, AI search engines can provide more accurate and relevant results, reducing the need for users to sift through irrelevant information.\n\nIn conclusion, the topic of Semantic Search and Natural Language Processing (NLP) in AI Search Engines is a hot and growing area of discussion and development. As we approach the year 2024, we can expect to see continued growth and innovation in this field, driven by the increasing demand for more intelligent and intuitive search experiences.”
}
]
bundle 2：
[
{
“text”: “As we look towards 2024, the topic of "AI-Powered Personalization in Search Engines" continues to be a popular theme in discussions and articles. With the rapid advancement of artificial intelligence (AI) technology, personalization has become a key trend in search engines. By leveraging machine learning algorithms, AI search engines can analyze user behavior, preferences, and context to create highly customized search experiences (Schmidt, 2023). This means that instead of delivering generic search results, AI-powered search engines can tailor the results to each individual user, providing more relevant and useful information.\n\nAccording to a report by Gartner (2022), AI-powered personalization in search engines is expected to drive a 25% increase in user engagement and satisfaction by 2024. Moreover, a survey by Econsultancy (2023) found that 86% of marketers believe that AI-powered personalization in search engines is critical for their business success.\n\nHowever, the rise of AI-powered personalization in search engines has also raised concerns about data privacy and security. As search engines collect more data about users to provide personalized results, there is a risk that this data could be misused or stolen (Smith, 2023). Therefore, it is important for search engine companies to have robust data protection policies in place to ensure the confidentiality and security of user data.\n\nIn conclusion, the trend of AI-powered personalization in search engines is set to continue in 2024 and beyond. By delivering more tailored and relevant search results, AI search engines can improve user engagement and satisfaction. However, it is crucial for search engine companies to address data privacy and security concerns to maintain user trust and confidence.\n\nReferences:\n\nGartner. (2022). Predicts 2023: CRM Customer Service and Support. Gartner.\n\nEconsultancy. (2”
}
]
i do have some programming exp.
but the aggragater is so difficult to use ,if you are official man. it is grateful for you to let the pm
know this…
To “combine” multiple bundles into a single structure, you’ll need to use an aggregator module.
Aggregators are modules that accumulate multiple bundles into one single bundle. An example of a commonly-used aggregator is the Array aggregator module. The next popular aggregator is the Text Aggregator which is very flexible and has applies to many use-cases (not just combining into a single plain text variable!)
There are other types of aggregator modules, click the below links to find out more:
- Array aggregator
- Text aggregator
- Numeric aggregator
- Table aggregator
- JSON aggregator
- CSV aggregator
- Archive (ZIP) aggregator
- other app-specific aggregators
Your aggregator’s source module in your case, should be the Iterator module, because that’s where the multiple bundles are being generated from.
Next time, please begin your question with your entire scenario.
samliew – request private consultation
Join the Make unofficial Discord server!
2 Likes
Ok heres a solution im pretty confident will work.
You will need to split this scenario up into 2 different scenarios. I’ll explain why at the end.
-
After you get your 3 different output texts, create a new Google Sheets Function (add a new row) then add those to new rows.
-
Open up a second scenario, you can either search for those rows, or if it’s more dynamic you can have it watch for those rows that were just added.
-
Add a text aggregator and set the value for the rows you either searched or watched for in step 2 along with the specific column. The result should just be 1 operation that’s 1 long text.
-
You can continue with the rest of your scenario after the text aggregator.
The reason you have to split it up is because Make goes through the whole scenario before going back and doing the the 2nd, 3rd bundles etc. So if you were to do the search in the same scenario and then aggregate it would still have different outputs.
Try it out and see if it works. Im confident that it will work, just unsure about the theory as to why lol.
Actually I think you can use an Array Aggregator, it just depends on the what you link as the source module. You want the source module to go as far back to the original as possible.
Example 1:
Here is an example of me linking the source module back to the original. The result is 1 operation an array that contains the results of the previous 3 operations.
Example 2:
Here is an example of me linking the source module just to the results I got, notice how I end up with multiple bundles which is not what we’re looking for.
@gstarwd I think you can probably do this in one scenario, your issue might be in what you’re linking as your source module.
Hi @gstarwd,
There are 3 operations in that Tools module because something that came before it ran 3 times because it produced three bundles (Maybe the Iterator #23?).
To handle this, you need to add some sort of Aggregator (like Text Aggregator, Array Aggregator, JSON Aggregator, etc…).
In your case, sounds like you might want a Text Aggregator.
Set the source of the Text Aggregator to whatever module is producing 3 bundles here.
This might not be the correct answer, though, because as @samliew pointed out, we need to see the entire scenario to give you a more accurate answer here.
If you change it along the way then you need to provide an updated version of it.
If you have API keys in your blueprint, there are also ways around that as well. If you have an API key in your blueprint, just change the text (and any other sensitive info) in the blueprint file before you post it.
2 Likes