Introduction:

Data Looksee is a GUI based data analysis tool. Users can utilize the features provided in the application to perform data mining operations and can extract meaningful information from raw data.

Application is designed to be used as End-User Driven User Interface; it facilitates the user to draw data concepts dynamically (also referred as Mind maps / Analysis). Analysis revolves around a node and its data configuration; we get unstructured data on each node to select and move to the next step or branch the concept. Ready to use datasets are loaded in the application and users have the flexibility to draw any concept from the pick and choose user interface.
User also has the ability to perform sentiments on unstructured data.

Login:

User must login to Data Looksee, by entering his login details:
• Customer Number: a 6 digit unique number for each customer.
• Username: username of the customer (email address).
• Password: password of the customer (case sensitive).

Account Recovery:

In case a user forgets his customer number or password, he can recover these details using his username, an email with details will be sent to his registered email address. In case a user forgets his username (email address) he needs to contact administrator. Steps:
1. On login page, click the link “Having trouble signing in?”
2. Select account recovery type: forgotten password or forgotten customer number.
3. Enter username.
4. Reply to secret question’s answer you saved in your user profile.
5. An email containing details for account recovery will be forwarded to your registered email address.

Toolbar:

Toolbar provides a set of features to facilitate user operations.

New Project [short key: ALT+N]

To create a new project click New Project button and select dataset from the list of datasets in which the new project is to be created.

Open Project [short key: ALT+O]

To open an already saved project click Open Project and select project to open. This menu also allows the user to open AutoSaved projects and Assigned Projects.

Save Project [short key: ALT+S]

To save a project, click Save Project button and enter project details: Name and Description. If you select overwrite option, a project that already exists with same name will be overwritten.

Save As Project [short key: ALT+X]

Save As allows you to save a copy of an existing project with a new name and description.

Export Project [short key: ALT+E]

To export a project, click on Project Export icon and select format to export (PDF or Excel).

Undo [short key: CTRL+Z]

To Undo last action performed on canvas, click Undo.

Redo [short key: CTRL+Y]

To Redo last Undo action, click redo.

Full Screen [short key: ALT+F]

Toggle between Full Screen view and normal view.

Logout [short key: ALT+L]

Exits application and logs out the user.

Rearrange Nodes [short key: ALT+R]

Rearrange Nodes button automatically arranges all nodes on canvas.

Reset Tab [short key: ALT+K]

Reset Tab clears current tab. All drawings are cleared.

Question & Answer [short key: ALT+Q]

Question & Answer allows user to create a mind map based on textual queries using natural language processing.

Settings:

User Profile:

User can update their profile using this feature.

Change Password:

User can change their password using this feature by providing their old password and new password.

User Settings:

Layout:

User can select their layout preference to draw mind map.

QA Settings

QA Settings allow users to set their preferences for QA results. User can select different features (Auto discover, clustering, Summary, Vocab-All) and different Charts to apply on these features.

Empty Node Selection Limits

Empty Node Selection Limits defines the number of results to be selected by default if user leaves a node empty.

Window Size

Window size defines a criteria to select terms. e.g. if window size is 4, and we look for nouns after a specific term, then all nouns that exist within 4 words window from that term will be selected. By default, this property is set to all, which means all terms will be selected irrespective of their distance.

Dataset loading preferences

User can define their preferences to load dataset.
If the user selects a default dataset, it will be loaded whenever user logs in.
If the user selects No-dataset, no dataset will be loaded.
If Last Used Dataset is selected, users last used dataset will be loaded on his next login.

Auto save time

If the selected project will be automatically saved after a user defined time. Auto saved projects help users to retrieve their unsaved work, in case they accidently logout.

Menu display time

Defines the time after which the menu appears on node hover.

Dialog Hide Time

Defines the time after which menu hides.

Dialog Appear

Defines the time to show a dialog to stop any request that’s taking too long to execute.

Post sentiment chart analysis

This setting allows the user to select how he wants to retrieve sentiment data in sentiment charts.

Sorting preferences

Allows user to select sorting preferences for terms in autocomplete menu.

Canvas:

Canvas is an area where the user can perform the analysis by drawing different mind maps in multiple tabs.


Canvas further divided into following areas:

1. Tab Management
2. Tab Menu
3. Zoom In Out
4. Autocomplete Menu
5. Node Hover Menu

Tab Management:

Add New Analysis:

User may add multiple tabs by clicking ‘Add New Analysis’ button from Tab bar.

Tab Menu:

A tab menu opens when the user clicks on down arrow icon inside tab and this icon get appears when the user focuses on the tab.


Tab menu further divided into following categories:

1. Window Size
2. Edit
3. Delete Tab
4. Duplicate Tab
5. Rearragne Tab

Window Size:

User may set term window size using the most relevant slider. By setting most relevant size, it finds the distance between given words/keywords, filter the results and highlight them so that we get the most realistic and relevant results.

For example to search for "food" and "dishes" within 10 words of each other in a document use the term size ‘10’.

Edit:

User may rename the tab text using the edit option.

Or

User can easily rename the tab by just double clicking on tab’s text.

Delete Tab:

User may remove focused tab by clicking on ‘Delete’ option from above menu.

Note: System always prompt a confirmation message before deleting any tab if the user forgot to save analysis.

Duplicate Tab:

Duplicate option enables the user to copy existing analysis in a separate tab.

Rearrange Tab:

Users allow to move tabs back and forth. Drag the tab using mouse and drops it at your desired location within tab bar.

Zoom In Out:

This feature enlarges everything on the screen except charts.

On the bottom right of the application screen, click the arrow to the right of the Change Zoom Level button or drag the slider right side.


Autocomplete Menu:

A menu/window is associated with every node on canvas. This menu helps the user to search data, make selections, remove selection and assist navigation in node.


Autocomplete Menu further divided into following categories:

1. Node Information
2. Data Selection
3. Top Bar Navigation Buttons
4. Remove Selection
5. Searching

Node Information:

The user may view node’s detail in separate ‘Node Info’ screen which appears when the user clicks on ‘Info’ icon from Top bar.

Items Selection:

User may view node’s detail in separate ‘Node Info’ screen which appears when user clicks on ‘Info’ icon from Top bar. Application allows to make single & multiple selection in nodes.

• User may select values from node by single click on check boxes. Whenever the user selects a value, it’s appeared in both the selection area and top bar.



• User may select multiple values using Top/Bottom window.


Top Bar Navigation Buttons:

Using the navigation buttons user may see incomplete selection from top bar.

Remove Selection:

User may remove selected values either from the top bar or the selection window.

Searching:

Application supports two types of searching.

1. Text Based Search
2. Quoted Search

Text Based Search:

User may search values by simply typing text in search field.

Quoted Search:

Using this feature user may search values having spaces among text.

Node Hover Menu:

There is a menu associated with every node which appears when the user gets the mouse over node. This menu helps the user to extend their analysis in multiple directions.


Menu further divided into following categories::

1. Vocabulary
2. Sentiments
3. Details
4. Clustering
5. Analysis
6. Related
7. Filter
8. Export

Vocabulary:

Vocabulary enables the user to filter data by parts of speech (Nouns, Verbs and Adjectives) to deliver visuals that tell the story instantly so user doesn’t have to struggle to interpret the data, but instead they understand it at first sight.

Vocabulary further divided into following categories:

1. Noun
2. Verb
3. Adjective
4. All
5. Auto Discover

Noun:

It shows the number of nouns appeared in documents against selected value.


In above image user selected term is ‘Doctor’ which appeared in any documents ‘895’ times then user selected Vocabulary > Noun which shows number of nouns e.g. ‘nurse: 44’ against ‘Doctor’ in any documents.

Verb:

It shows number of verbs appeared in documents against selected term.


In above image user selected term is ‘Doctor’ which appeared in any documents ‘895’ times then user selected Vocabulary > Verb which shows the number of verb e.g. ‘ask: 18’ against ‘Doctor’ in any documents.

Adjective:

It shows the number of adjectives appeared in documents against selected term.


In above image user selected term is ‘Doctor’ which appeared in any documents ‘895’ times then the user selected Vocabulary > Adjective which shows the number of verb e.g. ‘call: 16’ against ‘Doctor’ in any documents.

All (Noun, Verb & Adjective):

This node shows the number of ‘noun, verb & adjective’ appeared in documents against selected term.


In above image user selected term is ‘Doctor’ which appeared in any documents ‘895’ times then the user selected Vocabulary > All which shows number of ‘noun, verb & adjective’ against ‘Doctor’ in any documents.

Auto Discover:

This is a special kind of node which discovers and displays finding automatically on behalf of the root keyword; this auto discovery finds the things related to the selected keyword/term automatically.


In above image user selected term is ‘Doctor’ which appeared in any documents ‘895’ times then the user selected Vocabulary > Auto Discover which shows values of what is interesting against ‘Doctor’.

Sentiments:

It helps user to analyze sentiment and gives you the ability to see the true picture how those feelings evolve and change over time. It shows sentiment (positive, negative & neutral) results against selected term.

Sentiments further divided into following categories:

1. Sentiment Positive
2. Sentiment Negative
3. Sentiment Neutral
4. Sentiment Positive & Negative
5. Sentiment All

Sentiment Positive:

It shows positive results against selected terms. For instance, a user wanted to see positive sentences against any term then he/she will select Menu > Sentiments > Positive.


In above case the user selected ‘hospital’ term then draws positive node from Menu > Sentiments > Positive and finally dropped ‘Menu > Details > Sentences’ which shows positive sentences against ‘hospital’.

Sentiments Negative:

It shows negative results against selected terms. For instance if a user wanted to see negative sentences against any term then he/she will select Menu > Sentiments> Negative.


In above case user selected ‘hospital’ term then draw negative node from Menu > Sentiments > Negative and finally dropped ‘Menu > Details > Sentences’ which shows negative sentences against ‘hospital’.

Sentiments Neutral:

It shows neutral results against selected terms. For instance if a user wanted to see neutral sentences against any term then he/she will select Menu > Sentiments > Neutral.


In above case user selected ‘hospital’ term then draws neutral node from Menu > Sentiments > Neutral and finally dropped ‘Menu > Details > Sentences’ which shows neutral sentences against ‘hospital’.

Sentiments Positive & Negative:

It shows positive & negative results against selected terms. For instance if a user wanted to see positive & negative sentences against any term then he/she will select Menu > Sentiments > Pos | Neg.


In above case user selected ‘hospital’ term then draw Menu > Sentiments > Pos | Neg node and finally dropped ‘Menu > Details > Sentences’ which shows positive & negative sentences against ‘hospital’.

Sentiments All:

It shows positive, negative & neutral results against selected terms. For instance if a user wanted to see sentiment all sentences against any term then he/she will select Menu > Sentiments > All.


In above case the user selected ‘hospital’ term then draws Menu > Sentiments > All node and finally dropped ‘Menu > Details > Sentences’ which shows positive, negative & neutral sentences against ‘hospital’.

Details:

This section figures out detail level information of selected values at any time and any stage. User has the ability to see either clauses, sentences, document or summary of selected terms.

Detail section is further divided into following categories:

1. Documents
2. Sentences
3. Clauses
4. Summary

Documents:

Document is the combination of paragraphs which are stored in backend part of the application. If a user wanted to see the documents of any selected term then he/she will simply add documents node from Menu > Details > Documents.


In above image user first selected ‘hospital’ term from root node then draw Menu>Details>Documents node which shows documents against ‘hospital’.

Sentences:

A document is further divided into sentences. If a user wanted to see the sentences of any selected term then he/she will add the sentences node from Menu > Details > Sentences.


In above image user first selected ‘hospital’ term from root node then draw Menu > Details > Sentences node which shows sentences against ‘hospital’.

Clauses:

A sentence further subdivided into clauses. If user wanted to see the clauses of any selected term then he/she will add the clauses node from Menu > Details > Clauses.


In above image user first selected ‘hospital’ term from root node then draw Menu > Details > Clauses node which shows clauses against ‘hospital’.

Summary:

High level details of a document. If user wanted to see the summary of any selected term then he/she will add the summary node from Menu > Details > Summary.


In above image user first selected ‘hospital’ term from root node then draw Menu > Details > Summary node which shows summary against ‘hospital’.

Clustering:

Clustering facilitates ongoing analysis of content simultaneously analyze many documents and automatically extracting a set of phrases that best present the relationships between the documents. It dynamically creates topic categories across a set of content.

User may add the cluster node by follow the procedure below:

1. Add root node and select desired term.
2. Click Menu> Clustering > Cluster.



Above image shows clusters against ‘hospital’ term

Analysis:

Analysis part helps user to find problems and entities from unstructured data. Analysis part is further divided into two parts:

1. Problems
2. Entities

Problems:

Application extracts problematic parts/keywords from unstructured data and shows in a separate entity. For Instance if user interested to see the problems against any selected term then he/she will perform the following steps:

1. Add root node and select desired term.
2. Click Menu> Analysis > Problems


Above image shows problems against ‘hospital’ keyword.

Entities:

Big data comprises of big text which is lying hidden within that text is very valuable information, unable to be accessed unless read manually. This hidden data often comes in the form of entities—names, places, dates, and other words and phrases that establish the real meaning of the text.

For Instance if user interested to see the entities against any selected term then he/she will perform the following steps:

1. Add root node and select desired term.
2. Click Menu> Analysis > Entities


Above image shows entities against ‘food’ keyword.

Related:

Application instantly discover the related tone of important conversations and trail the impact of your social planning over time with an ability to drill down into discussions and analyze driving force behind positive and negative sentiments.

User has following options in related:

1. Clauses
2. Sentences
3. Documents

Clauses:

User can add related clauses as follow:

1. Add root node and select desired term.
2. Click Menu> Related > Clauses.

Sentences:

User can add related sentences as follow:

1. Add root node and select desired term.
2. Click Menu> Related > Sentences.

Documents:

User can add related documents as follow:

1. Add root node and select desired term.
2. Click Menu> Related > Documents.

Structure Filter:

Analyzes data depending on your specific business question. With total sustainability of each analysis, you can filter, measure and customized categories trend to evaluate your targeted advertising campaigns, promotions, and product launch effect that matters to your business.

User may add structure filter from Menu > Filter > Structure Filter



If user wanted to filter data on basis of structure filter then follow the steps below:

1. Add root node and select desired term.
2. Click Menu> Filter > Structure Filter.
3. In structure filter add ‘Rating=4” filter and press apply button.
4. Add noun node from Menu> Vocabulary > Noun ahead structure filter node.


In above image, Noun comes against rating 4.

Export:

Visualize your data any way you want. Easy to understand charts, grid and graphs highlights any insights that otherwise lay hidden in your data. Application uses state of the art visualizations.


Following are the list of available visualizations exist in system:

1. Bar Chart
2. OS Chart (Ordered Squares Chart)
3. SM Chart (Sentiment Matrix Chart)
4. Treemap Chart
5. Sankey Chart
6. Text Chart
7. Grid
8. Pdf
9. Excel

Bar Chart:

A visual presentation of precise information. In Vertical Bar Charts, bars are plotted vertically.

User may add bar chart by following way:

1. Add root node and select desired term.
2. Click Menu> Vocabulary > Noun.
3. On noun node click Menu> Export > Chart > Bar Chart.
4. Click on Bar Chart node.

OS Chart (Ordered Squares Chart):

An Ordered Square Chart display data points in ordered way according to their weightage. Datapoint with high weightage appears first and then low weightage datapoints. Datapoints labels display as callouts and its weightage display as square size. Inside of square we can display its weightage.

User may add os chart by following way:

1. Add root node and select desired term.
2. Click Menu> Vocabulary > Noun.
3. On noun node click Menu> Export > Chart > OS Chart.
4. Click on OS Chart node.

SM Chart (Sentiment Matrix Chart):

This chart displays sentiment related information column wise and detail row wise. Cross sectional details are displayed inside the bar.

User may add sm chart by following way:

1. Add root node and select desired term.
2. Click Menu> Vocabulary > Noun.
3. On noun node click Menu> Export > Chart > SM Chart.
4. Click on SM Chart node.

Treemap Chart (non-sentiment):

Tree maps Chart is a Chart where set of nested rectangles presented in hierarchical (tree-structured) data. Each branch of the tree is granted a rectangle, which is then tiled with smaller rectangles representing sub-branches. Rectangles has a zone area proportional to a specified dimension on the data. When the color and size dimensions are correlated in some way with the tree structure, one can often effortlessly see patterns that would be difficult to spot in other ways.

User may add treemap chart by following way:

1. Add root node and select desired term.
2. Click Menu> Vocabulary > Noun.
3. On noun node click Menu> Export > Chart > Treemap Chart.
4. Click on Treemap Chart node.

Treemap Chart (sentiment):

If the second level nodes in data represent sentiment we call it TreeMap Sentiment Chart. This chart is used to display hierarchical sentiment data with three levels only.

User may add treemap chart (sentiment) by following way:

1. Add root node and select desired term.
2. Click Menu>Sentiment>Pso | Neg.
3. Add vocabulary node ahead sentiment Menu>Vocabulary > Auto Discover.
4. On noun node click Menu> Export > Chart > Treemap Chart.
5. Click on Treemap Chart node.

Sankey Chart (non-sentiment):

A Sankey Chart is a visualization used to illustrate a flow from one set of values to another. Links are created through nodes. Sankey charts are most suitable when you want to show a many-to-many mapping between two domains (e.g., schools and majors) or multiple way paths through a set of stages.

User may add sankey chart by following way:

1. Add root node and select desired term.
2. Click Menu> Vocabulary > Noun.
3. On noun node click Menu> Export > Chart > Sankey Chart.
4. Click on Sankey Chart node.

Sankey Chart (sentiment):

A Sankey Sentiment Chart is a visualization used to illustrate a flow from one set of values to another. Links are created through nodes. In Sankey Sentiment Chart second levels nodes are considered sentiment nodes. Sankey Sentiment Chart is a three level chart only.

User may add sankey sentiment chart (sentiment) by following way:

1. Add root node and select desired term.
2. Click Menu>Sentiment>Pso | Neg.
3. Add vocabulary node ahead sentiment Menu>Vocabulary > Auto Discover.
4. On noun node click Menu> Export > Chart > Sankey Chart.
5. Click on Sankey Chart node.

Text Chart:

Text chart is normally used to show text information that’s why text chart node only visible ahead details node.

User may add text chart by following way:

1. Add root node and select desired term.
2. Click Menu> Details > Sentences.
3. On noun node click Menu> Export > Chart > Text Chart.
4. Click on Text Chart node.

Grid:

Grid is a default chart which can be attached to any node. Grid can be loaded with heavy data and provide sorting and paging facilities.

User may sort data on the basis of details and rank. Grid has the ability to show sentiment based results.

User may add grid by following way:

1. Add root node and select desired term.
2. Click Menu> Vocabulary > Noun.
3. On noun node click Menu> Export > Grid.
4. Click on Grid node.


Sentiment Based Result:

Pdf/Excel Export:

User may export any specific node’s data using pdf & excel widget.

Following is the way to add pdf/excel widget:

1. Add root node and select desired term.
2. Click Menu> Vocabulary > Auto Discover.
3. On auto discover node click Menu> Export > Pdf/Excel.
4. Click on PDF or Excel node.


Whenever user clicks on either pdf or excel node; auto discover data will be exported to the relevant application.

Widget Toolbar:

There is toolbar embed in every widget which encourages the user to export, visualize & refresh data from widget.