In this day and age, everyone has a blog.
Niche blogs are becoming very important. Today I wanted to write about a niche: Analytics and Bayesian techniques. Bayesian techniques power the best possible kind of analytics. If the area of research is qualitative then well that could be a little different. Bayesian classifiers and NBC blogs are rare. They are written by Postdocs and above. I surveyed around 40 niche’ blogs before coming to a conclusion.
Here is a pattern I see in them
- An NBC with real data in a blog will not exist unless the data are from public sources.
- The comments on these blogs are from a handful of the friends of the blogger.
- They are less likely to teach the real tricks involved.
- They don’t seem to compare the other techniques.
I am merely using NBCs as an example.
Here is why we don’t find good information on an NBC and the like in a niche blog:
Citations : Required but are time consuming
The audience is small.
No one wants to point out a conceptual folly at that level.
So why blog about an NBC at all? To create a living breathing space that showcases one’s facility with large data sets, understanding of prior and posterior probabilities and to show predictive power in their technique. Who are they aimed at? They are aimed at the next project manager or a multinational company that needs analytics based decisions.
There is another very very important side to it I: Can you figure out why such a blog is useful?
I worked on a project that needed me to build a good NBC when i was at school. It was an ambitious project and i did it.
I then found out that I had very little right over it. The Prof from the other department said i owned it but my own department wanted to share it. I finished the code but took away the data set that would prove an important hypothesis and provide evidence of good predictive power in a multidimensional data set. Many people were irate with me. I told them about the contract I had with the school. I did not generate the data with the schools resources. I did that on my own when I had no funding for the summer.
After that I was really cautious with my code.
Here is something all students need to know: If the code has been created or for that matter any kind of work has been done, with funding, then the student is not the sole owner the of the work. However, the student has to be very careful in documenting the steps and gathering evidence of contributions to it.
Here are some very unfortunate cases of students losing so much:
Here is something that will interest you!
After plagiarism allegations, prof. was named dean
Some tenured faculty members in English dept. remain troubled by U.’s handling of the matter
Published: September 24, 1997TORONTO, Sept. 23— Ruling that a professor at the University of Ottawa had plagiarized an M.B.A. student’s paper, an Ontario court has ordered the university and the professor to pay the former student $7,500 in damages and to reimburse legal costs.
UNL Student Alleges Plagiarism, Sues Former Profs
A former University of Nebraska, Lincoln doctoral student, who learned her first computer programming language at the age of 11, has filed suit against several of her former professors alleging that she was not allowed to finish her Ph.D. after complaining of plagiarism by her adviser.
“I don’t want to lose the five years of my life that I’ve spent doing this research,” Padmapriya Ashokkumar, a Chennai, India native, told India-West. “I’ve made an emotional as well as a financial investment, and I hope that at the end, something good will come out of all of this.”
Ashokkumar came to UNL in 2002, specifically to work on her Ph.D. with Scott Henninger, whose work she had read about while still in India. The Indian American student came to the university a month before classes started to get a jumpstart on her project, which was in the field of computer usability design; Henninger served as her adviser on the research.
But within a short period of time, Ashokkumar found herself frustrated by Henninger’s consistent inaction regarding her projects, she said. The young student tried to get a paper published in an academic journal, but Henninger delayed submitting it, challenging the quality of Ashokkumar’s work, she alleged.
Ashokkumar, however, does not believe Henninger was racially biased, merely that he was a procrastinator.
“It wasn’t just me,” Ashokkumar told India-West. “I saw that he had the same pattern with several of his students and I realized that if I wanted to get ahead, I would have to switch advisers.”
In 2007, Ashokkumar told Henninger she planned to switch advisers. Henninger allegedly became upset with her, according to court papers.
Henninger then submitted Ashokkumar’s paper for publication, removing her name from the work and identifying himself as the sole author; he did, however, credit Ashokkumar in the acknowledgements. Henninger protested when Ashokkumar – with her new advisers Sebastian Elbaum and Lorin Hochstein – submitted a similar work to a software engineering conference.
Later that year, Ashokkumar filed a plagiarism complaint against Henninger, who filed a counter-complaint against his former student. A UNL committee reviewed both complaints and found in Ashokkumar’s favor, noting that the credit in the body of the paper was not enough to acknowledge the extent of Ashokkumar’s research.
But despite the committee’s ruling, new department chair Steve Goddard would not allow Ashokkumar to continue with her work in the same field and advised her to start in an entirely new area. Moreover, other professors in the department would not advise Ashokkumar on the research she had spent five years pursuing, she alleged.
“I would have had to start all over again, wasting five years of my life,” said Ashokkumar. “Even when I considered restarting, I did not receive a positive response from the department.”
Ashokkumar – who now works for a private software company in Austin, Texas, allegedly for far less pay than she would have received if she had had a Ph.D. – filed suit this April against Henninger, Elbaum, Hochstein, and Goddard. Kimberly Espy, former associate vice chancellor for research at UNL; and Prem Paul, vice chancellor for research and dean of Graduate Studies, are also named in the suit.
Reached at his UNL office, Goddard – through his administrative assistant – declined to comment for this story. Henninger, who has since left UNL, could not be reached.
Gene Summerlin, who is representing Ashokkumar in her lawsuit, told India-West that his client had been sucked into “an academic turf war.”
“Dr. Henninger was working in a very new area of computer programming and was attempting to claim proprietary rights to all research in that area,” Summerlin said.
There were several attempts to settle the matter out of court, but Henninger insisted on sitting on Ashokkumar’s thesis committee, he added.
“Padma was done with all her coursework and had only to defend her dissertation. The university had an obligation to restore her standing, but instead effectively terminated her work on her Ph.D.,” stated Summerlin.
The lawsuit seeks monetary damages for the five years that Ashokkumar has worked at salaries far less than she would have received as a Ph.D. holder. Summerlin believes this to be about $20,000 a year. The suit also seeks to restore Ashokkumar’s status at UNL and allow her to defend her dissertation and receive her Ph.D. based on the work she has already completed.
With any outcome, Ashokkumar said she is determined to receive her Ph.D.
“I have loved technology since I was in 8th grade with my little home computer, and I have the skills to do the research to work in the field,” she asserted to India-West.
Author’s comment: Speaking of the statistics related to blogs, could a person like this start a niche blog? I bet she would be beyond popular. If she were to have an opinion on say security patterns, uncertainty quantification etc. I would listen.