Differentiate a type using text mining/ machine learning with python or R
- or -
Post a project like this1708
$27/hr
- Posted:
- Proposals: 16
- Remote
- #2462735
- Awarded
Web Development,Desktop & Android Development , WordPress, Logo Designing, Graphic Designing, Scrapping, Data Science,
Rawalpindi
1382310258917028468532843011284274828150062797937278967927374492687838386347
Description
Experience Level: Intermediate
This is not a typical sentimental analysis as I saw on the web.
Three are three columns, ID, type_Code, and description. Type_Code (second column) is all “sinking”, what I want to do is to classify the type (column 2) for each ID based on description (column 3) into two types, basically, two types of “sinking”, one is “regular” sinking, mainly due to nature, weather, for example, key words may include rain, storm, wind, wave, flood, tide, no recent repairs, but not limited to these. Unknown reasons would also be classified as this regular
The other type is “bad” sinking: this one involves other things or people, or due to operators’ error, for example, hitting some objects, crash with other boats, wave/ wake caused by other boat, key words may include: involved in accident, hit the rock, hit the submerged object, hit the sso, crash, grounding, but not limited to these.
For example, ID 25546854 with description (column 3) has no obvious reason for the sink, then the type (column 2) of ID 25546854 is being classified as regular sinking.
ID 94620353 with description (column 3) has key words like Strong Winds/Bad Weather, No Known Cause Of The Boat, Prior To Placing It In Water With No Problems. Even these cell has some key words like “rock”, but rock is not the main reason. Therefore the type of ID 94620353 is being classified as regular sinking,
Some notices:
1. The following words alone can’t be used as keywords to classify the “sinking” type, even it has high frequency
Insureds(insd), water, sink, boat, feet, inches, foot, sinking, propeller, shaft, cabin, bilge, pump, number-like words (3, five, 54$, etc.)
2. Notice there might be negative words in front of key words, for example, did not hit, didn’t hit,
3. “other boat” not necessarily imply the “bad” type of sinking, since it may be “other boat helped…”
Three are three columns, ID, type_Code, and description. Type_Code (second column) is all “sinking”, what I want to do is to classify the type (column 2) for each ID based on description (column 3) into two types, basically, two types of “sinking”, one is “regular” sinking, mainly due to nature, weather, for example, key words may include rain, storm, wind, wave, flood, tide, no recent repairs, but not limited to these. Unknown reasons would also be classified as this regular
The other type is “bad” sinking: this one involves other things or people, or due to operators’ error, for example, hitting some objects, crash with other boats, wave/ wake caused by other boat, key words may include: involved in accident, hit the rock, hit the submerged object, hit the sso, crash, grounding, but not limited to these.
For example, ID 25546854 with description (column 3) has no obvious reason for the sink, then the type (column 2) of ID 25546854 is being classified as regular sinking.
ID 94620353 with description (column 3) has key words like Strong Winds/Bad Weather, No Known Cause Of The Boat, Prior To Placing It In Water With No Problems. Even these cell has some key words like “rock”, but rock is not the main reason. Therefore the type of ID 94620353 is being classified as regular sinking,
Some notices:
1. The following words alone can’t be used as keywords to classify the “sinking” type, even it has high frequency
Insureds(insd), water, sink, boat, feet, inches, foot, sinking, propeller, shaft, cabin, bilge, pump, number-like words (3, five, 54$, etc.)
2. Notice there might be negative words in front of key words, for example, did not hit, didn’t hit,
3. “other boat” not necessarily imply the “bad” type of sinking, since it may be “other boat helped…”
Rebecca M.
100% (1)Projects Completed
1
Freelancers worked with
1
Projects awarded
100%
Last project
22 Jul 2019
United States
New Proposal
Login to your account and send a proposal now to get this project.
Log inClarification Board Ask a Question
-
There are no clarification messages.
We collect cookies to enable the proper functioning and security of our website, and to enhance your experience. By clicking on 'Accept All Cookies', you consent to the use of these cookies. You can change your 'Cookies Settings' at any time. For more information, please read ourCookie Policy
Cookie Settings
Accept All Cookies