Pluaris User Guide
Pluaris automatically reads and analyzes textual data for you, saving you time, increasing your productivity, and making it easier for you to find the information you need.
Introduction
Pluaris can take in textual data from news articles, websites, a wide variety of document types, and your own notes, all of which become Reader section in the system. The AI engine automatically processes each article as it is uploaded, creating a summary, highlighting key drivers, categorizing the content, and identifying specific entities, such as companies, people, locations and more.
Pluaris can be used to create an annotated news feed on your topics of interest. This news feed is automatically updated, and you can quickly scan the results at regular intervals (daily, weekly, or whatever fits your schedule) for the most relevant content. See Topic Monitoring for more information on how to set this up.
Pluaris maintains a searchable knowledge base of the information you have uploaded to the system. This knowledge base is not just keyword searchable however – the metadata resulting from the AI engine analysis of each article facilitates intricate drill-down capabilities that reveal hidden connections and unexpected insights, and a granularity of search that provides high specificity.
This guide will walk you through the key steps to set up and use Pluaris:
- Inputs: Getting Data into Pluaris covers the ways in which you can get data into Pluaris so that the AI engine can automatically analyze and store that information for you to scan and search.
- My Memory: Exploring Your Knowledge Base will show you the different options for searching through the data that is in your Pluaris database.
- Notebook: Synthesize and Summarize details how to create notes based on the insights and information gleaned from your Pluaris database.
- Sharing shows you how to share articles and notes with your team.
Note that throughout the Pluaris application you can find helpful information and tips by clicking the Information icons.
Let’s get started!
Inputs: Getting Data into Pluaris
There are two primary methods of getting data into Pluaris, located under Inputs in the left-hand navigation menu:
- Quick Upload allows manual upload of websites, text or files.
- Topic Monitoring monitors web news sources for information related to your topics of interest.
Every website, text, file or news article that is uploaded to Pluaris becomes an Article in your database. Each of these articles are immediately processed by the Pluaris AI engine, automatically generating summaries, and identifying drivers, categories, and specific entities. My Memory: Exploring Your Knowledge Base will show you how you can use this information to quickly scan articles in your feed, generate new insights from the articles in your database, and find the information you need.
Note that there are actually two other ways in which data can get into your system:
- Articles shared with you by someone else on your team: this feature is only available to users who are part of a larger organization, and is discussed in detail under the Sharing section.
- Custom Integrations: Pluaris can be integrated with various enterprise tools (cloud drives, message boards, etc.) to create automated feeds. Contact Nowigence for more information.
Quick Upload
Quick Upload allows manual upload of websites, text or files. To begin, click the Quick Upload link in the left-hand navigation menu. There are three types of content that you can upload:
URL: To upload the content of a website, simply copy the URL and paste it in the “Enter a URL” field.
TEXT: To upload text you can simply type in the “Enter Your Content Here” box, or copy text from another source and paste it here. You can also give the article that is created from this text a title in the “Enter a Title” field.
FILE: To upload a file, click the Choose File button. This will open up a window to search for a file on your device. Select the file you want and click Open to return to Pluaris with the file of interest now selected in the Quick Upload window. Acceptable file types include PDF, DOCX, TXT, PNG, JPG, JPEG, and TIF.
Before uploading, you can optionally
- Label the incoming article with a Relationship by clicking in the Relationship field. This can be used to facilitate future searches, as outlined in the Filters section.
- Add your own custom Tag to the article, selected from the current list of Tags, or by creating a new Tag using the Plus button. Again, this information can be used to facilitate future searches, as outlined in the Filters section. Tags are managed under Topic Monitoring.
- Share this article immediately with members of your organization by selecting Share and then selecting the users with whom you want to share the article. For more information, see the Sharing section.
When you are ready to upload, click Upload to prompt Pluaris to pull the content into your knowledge base. To cancel and close the Quick Upload window, click Nevermind.
Topic Monitoring
Topic Monitoring allows you to monitor web news sources for information related to your topics of interest. You simply enter keywords related to the topics you want to monitor and save them in your profile. Pluaris then automatically runs a daily search of those topics against over 170k web news sources, fetching the top matches, analyzing them with our AI engine, and storing them in your system. You can then easily scan your news feed in the Reader view, reviewing summaries and key drivers, and can monitor trends and statistics in the Control panel view.
Note that news articles retrieved via Topic Monitoring are automatically deleted from your system after 30 days. However, you can easily move a news article to your knowledge base for permanent storage – see Reader Section for more information on how to do this.
To get started, click Topic Monitoring in the left-hand navigation menu:
In the field that is labeled “Enter a topic to monitor”, type a keyword or keywords that describe a topic of interest to you (e.g., Natural Language Processing).
Optionally, you can add keywords that you want excluded from your search in the field labeled “Add exclude topics”. This can help avoid extraneous results.
Finally, select a Relationship by clicking in the field labeled “Select how this topic relates to you”. This Relationship label will be applied to all future news articles found under this topic, and can be used to facilitate future searches, as outlined in the Filters section. If you are not selecting any relationship then by default it will be set to ‘Other’ as that is the default value.
You can see an example of the types of news articles a particular topic (along with any excluded words) will retrieve by clicking Show Preview. This will immediately retrieve up to five articles based on this topic, and display them as follows:
Use Show Preview and the information provided under Tips to hone your search results. When you’re ready to save it, click Add to Profile. Your new topic will be listed in the table below.
From your existing topic list, you can perform the following actions on an individual topic:
- Click the Menu icon to show a list of the most recent articles that have been added to your system based on this topic.
- Click the Edit icon to edit this topic. The topic will open in the upper menu to allow modifications.
- Click Trashcan icon to delete this topic.
As your topic list grows, you can use the search functionality in the field labeled “Search Your Topics!” to find the topics you want.
Tags
In the Topic list table, you can also manage your custom Tags. Click on the Article Tags icon to show your current list of Tags.
You can edit or delete existing tags, or create a new tag using the Plus button. Tags can be applied to articles in various ways, and can then be used to facilitate future searches, as outlined in the Filters section.
My Memory: Exploring Your Knowledge Base
The articles uploaded to your database via the various input options, along with their associated annotations automatically generated by the Pluaris AI engine upon upload, form a personal knowledge base of information on topics relevant to you. The pages in My Memory allow you to explore that knowledge base, whether it be monitoring your incoming data feed (such as the news feed articles generated by Topic Monitoring) or drilling down into your entire database to rediscover earlier information, make connections and find insights.
The Control panel(Dashboard) and Reader(Articles) views, in combination with the options available via Filters, facilitate this deep searching of your knowledge base.
Ask Me can quickly extract a specific answer to any natural language question you want to ask of your knowledge base.
These options are all located under My Memory in the left-hand navigation menu:
Control Panel ( previously the Dashboard )
The Control Panel provides a quantitative, graphical view of the annotations produced by the Pluaris AI engine. The Control Panel results represent articles uploaded to your database over the past 15 days. These results can be further filtered down by using the Filters menu, as described in the Filters section.
The following graphics can be found on the Control Panel:
This graphic includes:
- NEWS, which indicates the number news articles.
- RISKS and OPPORTUNITIES, which indicate the number articles labeled as a potential risk or opportunity, respectively (see the Risks and Opportunities graphic below for more information on this).
- PUBLIC UPLOADS, which indicates the number of articles that have been shared with or by you.
- PRIVATE UPLOADS, which indicates the number of articles that have been uploaded privately by you.
The percent change for each category indicates the increase (green) or decrease (red) compared to the previous 15 days.
This graphic shows specific entities found by the Pluaris AI system in the past 15 days of articles. The categories of entities for which Pluaris searches include Company, Product, Person and Location. The top 5 most commonly found entities are shown, along with the number of articles in which each appeared. Clicking on a specific entity (such as Facebook above, listed under Company) allows you to filter the results further by clicking Add to Filter (as described in the Filters section), and to give feedback to improve future search results (as described in the User Feedback section).
This graphic shows the number of articles mentioning specific countries in the past 15 days. Hover your cursor over a country to see a specific count and click on that country to filter the results further by clicking Add to Filter (as described in the Filters section). Click Cities to display the top 5 cities by number of articles mentioning them.
This graphic shows a list of articles in the past 15 days labeled as either Risk or Opportunity. Articles that are not identified as either a Risk or Opportunity are labeled as Informational. Click Risks, Opportunities or Informational at the top to switch which category you are viewing.
More information on how these labels are determined by the system can be found in the Risks and Opportunities section. The Plus button can be used to add text directly to a note (see the Notebook: Synthesize and Summarize section). Click More in the bottom left to open a more detailed view of the articles, including a Summary and associated Drivers (for more information on how these are produced, see the Summaries and Drivers sections, respectively).
This graphic shows a pie chart of the top 5 Categories found in the articles from the past 15 days. Clicking on a specific category (such as Innovation in the image above) allows you to filter the results further by clicking Add to Filter (as described in the Filters section). More information on how these various categories are determined can be found in the Categories section.
This graphic shows a bar chart of the top 5 Drivers found in the articles from the past 15 days. Clicking on a specific driver (such as Product Development in the image above) allows you to filter the results further by clicking Add to Filter (as described in the Filters section). More information on how these various drivers are determined can be found in the Drivers section.
Reader ( previously ‘Articles’ )
The Reader page provides a scrollable list of the articles in your database, starting with the most recent additions, making it easy to quickly scan your news feed. Using the filtering options described in the Filters section, you can explore your entire knowledge base by further narrowing down the articles shown in this view to get to the information you need to find. At the top of the Reader page, the system will indicate how many articles are currently being shown, as well as the total number of articles in the database (e.g., as shown below, “Retrieved 28 of 74 articles” indicates that 28 articles are being shown of the 74 total articles in this database).
Each article appears as a small card, which includes the title of the article, a peak at the first part of the Summary, and a label in the upper right corner indicating the source of the article. In the case of articles from the web, clicking on the title of the article opens the web site associated with that article in a new tab. For other articles (such as files or notes), a window with the full article text opens up.
By clicking More, the article card expands to show the various annotations associated with this article that were created by the Pluaris AI engine.
A Summary of the article and list of Drivers associated with this article are shown. In the left column, you will find the Category, Related Articles in your database (click on the titles to open the article in a new tab), any Tags associated with this article, and the specific Companies, Products, People, Locations and any Other entities found by the AI engine. You can add an existing tag to this article, or create and add a new tag. Clicking the X next to an entity under Company(s), Product(s), People, Location(s), or Other allows you to give feedback to improve future search results (as described in the User Feedback section). Clicking an entity name copies that name to your clipboard, which you can then paste in a note in your Notebook, or into the topic field in Topic Monitoring to quickly create a new topic of interest to monitor, or into any external application you want to use.
The three dots menu in the lower right of the un-expanded version of the article card allows you to share this article (see the Sharing section), delete it from your database, or you can select Add to My Memory in the case of news feed items to make that article part of your permanent knowledge base.
Filters
As noted in the Control Panel(Dashboard) and Reader(Articles) sections, Filters allows you to drill down into your knowledge base to find the information you need, providing an incredibly powerful granularity of search, packaged in an intuitive, graphical interaction. NOTE: Filters, including the keyword search filter, only apply to the Control panel and Reader views. Ask Me always searches against your entire knowledge base, regardless of the current Filter state.
Upon opening the Filters menu, at the top you will see:
- Reset will quickly reset the filter state back to having no filters selected.
- Clicking Save opens up a text box in which you can give the current filter state a name and then click + to save that filter state.
- Saved Filters opens up a list of filter states that you have previously saved – selecting any of these will immediately return you to that state.
The rest of the options available under the Filters menu allow you to filter on a specific label within that category.
- Topics corresponds to the topics you have setup in Topic Monitoring.
- Relationships corresponds to relationships that have been applied to articles. These relationships are optionally set in your Topic Monitoring topics and when you use Quick Upload.
- Tags can be applied manually to articles and are managed under Topic Monitoring.
- Data Sources relate to how the article entered your system.
- Categories correspond to the business relevant categories Pluaris uses to label articles, as described in Categories.
- Drivers are related to the intent or reason behind a decision, as described in Drivers.
- Outcomes can be either Risks or Opportunities, or else can be simply Informational, as described in Risks and Opportunities.
- Countries corresponds to the specific countries found in the articles and matches up with the Geography graphic in the Control panel.
- Company, Product, Person and Location are entities found in the articles by Pluaris and match up with those shown in the Trending graphic in the Control panel.
In addition to the options listed under the Filters menu, you can also perform a keyword search using the keyword filter bar at the top of the page in the Control panel or Reader view. Simply type in the word or words on which you wish to filter, and click the Search icon.
Note that when the Filters menu is closed, you can still see if you have any filters currently selected, as the Filters icon will have a small blue circle next to it with a number indicating how many filters are currently being applied:
In the Control panel view, you can quickly add new filters to your filter state by clicking on specific labels in the various graphics. Below are two examples.
After clicking on Innovation in the pie chart, the below box opens up. Simply click Add to Filters, and Innovation will immediately be selected within Categories as part of your filter state.
Similarly, in the Trending section you can click a specific entity (in this case, Twitter was selected under Company). Again, just click Add to Filters to immediately make that part of your filter state.
Ask Me
The Ask Me view allows you to ask specific, detailed questions of your knowledge base. Pluaris will attempt to find a specific answer and will also suggest other related articles from your knowledge base.
In the example below, the following question was asked of a Pluaris account being used to track artificial intelligence news:
Pluaris returned the following results from the user’s knowledge base:
The first part (Suggested Answer from your Memory) is the most specific piece of information that Pluaris found, with a link to view that article. The second part (Suggested Reading from your Memory) shows related articles with links.
Pluaris also goes out to the web at the same time to bring back the best answer and related articles outside of your knowledge base. Here are the results returned for this search:
Notebook: Synthesize and Summarize
In the Notebook, you can create your own notes, allowing you to easily pull together the insights and information you find as you explore your knowledge base. You can create and edit notes and add content from articles with a single click. Your notes are also automatically turned into articles in your knowledge base, so that they are searchable in all the same ways that news articles and uploads are. Notes can also be shared with others in your organization – for more on this, see the Sharing section.
Creating and Editing Notes
Click Notebook in the left menu to open an overlay window showing your notes. You can use the Sort by and Note Name Search… features at the top of the Notebook to help you find existing notes. To create a new note, click:
This will open the Note editor, which allows you to name the note, start typing content into it, and format it using the menu near the top:
The menu immediately above that looks like this:
Which, allows you to toggle the window to full screen, share the note with others in your organization, download the note as a PDF, or delete the note.
In addition to typing in your own content, you can also add content to your note anywhere you see the Plus button. To do this, you need to first activate the note by simply clicking Activate next to that note in the Notebook list view:
You can have more than one note active at a time. Clicking on the Plus button will send that content to all active notes.
Viewing Note Articles
Every note that you create in the Notebook also generates an article within your knowledge base. Here is an example article card in the Reader view resulting from a note:
The Pluaris AI engine analyzes the note in the same way it does any other article entering the system, and as a result the note article is searchable in all the same ways as other articles. In addition, the article associated with the note will be updated each time the note is edited and saved.
Sharing
The sharing functionality in Pluaris allows you to share interesting articles (including those created from your notes) with others in your organization.
Sharing Articles
Articles can be shared in two different ways. In the Reader view, click the three dots in the lower right corner of an un-expanded article card. Share will be one of the options available in this menu. Clicking Share will open a dialog box with a list of the other users who are part of your organization. You can select Send to All to automatically select all users or select any number of individual users. Then click Share to send a copy of the article. The article will then appear in the Reader view for those users with whom you have shared it, and they will also have full access to the same annotations in their Control panel.
Alternatively, articles can be shared immediately at the time of upload when using Quick Upload. From the Quick Upload window, simply select Share, which will open a similar list of the other users in your organization. If you select Private (rather than Share), the article will be uploaded only to your knowledge base. You can of course decide later to share that article from within the Reader view in the manner described immediately above.
Sharing Notes
Notes can be shared in two ways. The first is to open the note in the Notebook. Below the title of the note, click on Share icon to open a list of the other users in your organization with whom you can then share the note. Alternatively, find the article associated with the note (as described in the Viewing Note Articles section) and share the article as described above.
AI Engine
Pluaris is backed by state-of-the-art Natural Language Processing and other Artificial Intelligence algorithms in order to deliver the highest quality annotations of your data, enabling the granularity of search and unique insights that make Pluaris so powerful. These AI-based features include:
- Summaries
- Risks and Opportunities
- Categories
- Drivers
- User Feedback
Summaries
Pluaris pulls the key points from the article text to create a brief summary that you can quickly scan to get a basic understanding of the article, and to help you determine if you want to read it in more depth or skip over it to the next article on your list.
The Summary can be found in the article card in the Reader view by clicking More to see the expanded card:
Risks and Opportunities
A suggested Risk is defined as any article that might prevent you or your business from reaching its goals or objectives. A suggested Opportunity is defined as any article that could facilitate a positive or beneficial event. Articles labeled as Informational are less likely to trigger any action and may contain opinions.
The categorization into Risk or Opportunity is dependent on the relationship of the information. Relationships are assigned through Topic Monitoring or when uploading articles in Quick Upload. As an example, a company investing in new technology is typically an Opportunity. However, if that company is your competitor, it may be suggested as a Risk.
Categories
Articles in your knowledge base are automatically categorized into one of twelve business relevant categories.
Categories include:
- Awards and Recognitions
- Capacity
- Deals
- Executive Moves
- Innovation
- Investments
- Labor
- Legal
- Pricing
- Regulatory
- Sales
- Sociopolitical
Drivers
Drivers are defined as the intent or reason behind a decision being made. Articles within your knowledge base that contain sentences with a driver are automatically identified and categorized.
Driver categories include:
- Business Ethics
- Cost Reduction
- Customer Experience/Benefit
- Cyber Security
- Debt Reduction
- Digitization
- Funding/Growth
- Improper Maintenance/Negligence
- Labor Unrest
- Liability
- Logistics/Supply Chain
- Market/Customer Demand
- Market Expansion
- Product Development/Improvement
- Product Portfolio Optimization
- Productivity/Operational Efficiency
- Quality Assurance
- Raw Material Cost
- Regulatory
- Reputation
- Safety
- Shareholder Value
- Sociopolitical
- Sustainability
User Feedback
In the case of the Company, Product, Person and Location labels, you can correct errors made by the Pluaris AI engine in order to improve future results. This feedback can be given in two different places:
- In the Control panel view, in the Trending graphic, if you click on a specific entity under one of the four categories of Company, Product, Person or Location, a pop-up window will appear with the Add to Filters option described in the Filters This window also shows a link called Feedback, which you can click on to get the user feedback window shown below.
- In the Reader view, after expanding an article card using the More button, the four categories of Company, Product, Person and Location will be displayed on the left side under Insights. Clicking on the X next to a specific entity will open the same user feedback window shown below.
Here is the initial user feedback window:
Clicking into the Choose Reason drop down menu results in (the entity chosen in this case was NLP):
You can choose from the following reasons
- NLP Doesn’t Belong In Company: selecting this will allow you to reassign it to one of the other categories (Product, Person or Location in this case)
- I Just Don’t Want To See NLP: selecting this will simply remove it from your view
- Other: selecting this will remove it from your view as well, but also includes a field in which you can type specific feedback that our product team can use to improve our models for future releases.
When you are ready to send the feedback to the system, click the Submit.
