ISI Tales

Over.

2010-08-10T20:55:00.002+05:30

I am now a first year doctoral student at Cornell University, Ithaca, NY, USA.

Ergo, this blog has served its meager purpose and is now defunct.

There will not be any posts here from now on. I will update this post with a link to my new website and blogging platform once I set it up so anybody interested may continue to follow my chronicles there.

Citation for JYI Article in Current Science / Scopus Author Page

2010-07-13T12:46:00.003+05:30

I just received notification that my 2007 WTP article in JYI has been cited by an article in Current Science - which is India's most productive journal by impact factor (2008 IF - 0.774)

This is the second citation of the same article after the one in the World Bank report.

Also, my Scopus Author Page has been automatically generated. (I didn't know this was possible)

Ergo, pardon me if I sound a little kicked. =)

Done.

2010-05-16T23:20:00.003+05:30

Its done. The visa I mean. Now, I run after the health form and the other innumerable red tapes to slash at and hopefully in another two months (and a bit) I can see myself sniffing at the air outside JFK.

Paper Referred in World Bank Manual

2010-03-21T15:16:00.002+05:30

Back in 2007, I had published a paper in the Journal of Young Investigators, an undergraduate research journal.

I just received notification that this paper has been referred (hint: search for "Guha") in the World Bank Water and Sanitation Program Guidance Notes.

Click here to learn more about this World Bank program.

Phone Interview at the University of Washington, Seattle

2009-12-30T16:00:00.002+05:30

I am excited.

Last night, I received an email from the admissions counselor of the Information School at the University of Washington, Seattle.

I had applied for the PhD in Information Science at the ischool and it seems they have shortlisted my application for a phone interview on January 14, 2009.

Since this is my first phone interview, any tips, suggestions and even wacky ideas will be much appreciated.

I will post updates to this after the interview.

Talk delivered at the Indian Statistical Institute on Statistical Analysis of Internet Forum Usage.

2009-11-27T03:27:00.003+05:30

I delivered a talk on the application of statistical techniques or informetric tools to Web 2.0 applications. I mainly focused on the analysis and conclusions derived from quantitatively analyzing a forum dataset derived from the usage of a real life construction firm forum. This forum was actively used in supplementing the decision making process and in the solving of issues of a particular project of the construction firm.

Principles of informetrics were used to analyze the results and derive appropriate conclusions. An effort was also made to relate these findings with the classical bibliometrics laws of Bradford, Lotka and Zipf. These findings are also going to be published in a relevant conference pretty soon and I will put up those details as and when they come in.

The public announcement of the talk may be seen here.

The slides of the talk may also be found here.

Please go through these slides and comments are especially welcome.

Call For Papers and Participation for Workshop on Knowledge Transactions 2010

2009-11-27T02:37:00.004+05:30

The Call for Papers and Participation at the Workshop on Knowledge Transactions to be held at the Indian Statistical Institute, Bangalore Centre on 15th - 17th February 2010.

The appropriate link may be found here.

Here is a brief description of the scope of the workshop:

Knowledge is perceived and understood differently according to the context and purpose of tasks at hand. Libraries and Information centres have been involved in knowledge transactions in the sense of organization of collections, systems leading to context based indexing and techniques in handling and representing knowledge. In the digital information world representation models are increasingly based on traditional and theoretical foundations of knowledge transactions. Legacy methods based on principles of Knowledge Organization are now being adapted for Knowledge conceptualizations and representation and developing representation languages, knowledge architecture and structures. At the same time knowledge modeling using popular W3C-SKOS, NKOS and RDFs are being implemented especially to structure, organize and manage information resources. Documentation Research and Training Centre, Indian Statistical Institute, announces a seminar on 'Knowledge Transactions' to bring together expertise in information communities and active stakeholders in the domain.

The workshop will comprise of contributed papers, invited talks and demonstrations by resource persons, in areas pertaining to Knowledge Organization, Transactions and Management. The objective of the workshop is to provide a platform for discussions on latest developments in methods of knowledge transactions with a direct bearing and applications in Library and Information Science and services. The intended outcome is to enhance the knowledge and skills of the participants in methods of knowledge transactions. It is aimed to achieve the goal through:

Invited lectures
Contributed papers
Demonstrations
Discussion

The last date for sending in papers is 27th December 2009 and should be addressed to:

Dr. Devika P. Madalli,

Documentation Research and Training Centre, Indian Statistical Institute,

8th Mile, Mysore Road, Bangalore 560059

Email: devika@drtc.isibang.ac.in OR devikapm@yahoo.com.

Please participate or visit anyone who is even slightly interested in these topics. We will be waiting.

Paper Selected at Journal of Information Technology in Construction

2009-11-08T10:43:00.006+05:30

Edit:: December 28, 2009.

UPDATE 1: Online access to the paper may be found here.

UPDATE 2: The paper is available in the January 2010 issue of ITCon (Vol. 15, Pg. 86-107)

I recently had this paper of mine selected at the Journal of Information Technology in Construction. This is an open access journal indexed in various open access repositories like DOAJ etc.

The title of my paper is "Collaboration in a Web enabled Design Management System - A Case Study in Kolkata, India"

I used Web 2.0 data collected from my previous project in designing and developing a collaborative web based construction project management information system which was published as a book chapter in a symposium held in Dublin City University, Ireland. See this for more details.

I used the concepts of factor analysis and performed some degree of statistical social network analysis in order to analyze my data. Basically, I was looking to inquire into collaboration in online spaces. I wanted to see whether collaborative decision making in construction projects are different from any other similar projects. I found several interesting results which you can read in the paper.

It is currently in publication and I will update this post when it becomes available online.

Social Network Analysis Reading List

2009-08-20T18:23:00.004+05:30

This is the blog of Dr. Ralf Klamma, a professor of computer science at RWTH Aachen University in Germany.

Recently, I have been tweeting quite a bit and Prof. Klamma had added me since we are both interested in the same areas of analyzing social networks.

Since I am to start my MS thesis from September, I was voicing aloud my concerns that there was no integrated place on the internet from where I could get a comprehensive list of reading material for social network analysis (which itself is a vague and new topic !! :) )

Prof. Klamma tweeted back and gave me this link from his blog. This is a veritable treasure house of reading material and I could not have imagined someone painstakingly compiled this into such a large and useful list.

So, anyone who is interested in researching into social networks (of any kind) or just want to know something about it, please do visit the previous link. It is proving to be invaluable for me.

Paper Selected for Oral Presentation at ICSTE 2009

2009-06-15T06:51:00.003+05:30

I am back on this blog after a hiatus. Not that I didn't have anything to say but I wanted to say something worthwhile.

So, I sent in this paper for ICSTE 2009 and I got accepted. The paper is going to be indexed in ISI and published by World Scientific Press so I am excited. This means that (barring my indexed paper in IEEE Digital Library) I might have my first positive search result in ISI Web of Science and the Ei Compendex.

The full text of the paper may be seen here and the official acceptance letter may be found here.

Unfortunately, the dates of the conference have been pushed back from late June to late July. I have my 2nd semester final examinations in ISI then. This is why I will be unable to go for the conference. However, the administration has assured me that since I have paid the registration fees, my paper will be indexed and the conference proceedings and other materials will be sent to me.

On another note, a careful perusal of the conference letter reveals that 390 papers were selected out of 1305 submissions. That is an acceptance rate of 29.85% which is pretty decent.

Photos of SIRARM

2009-03-27T04:48:00.002+05:30

The photographs of the Seminar on Spatial Information Retrieval, Analysis, Reasoning and Modelling (SIRARM) are up on Photobucket in a slideshow format.

I might be found in a couple of photos but in case anybody is interested, here are the following slideshow links:

SIRARM Slideshow 1

SIRARM Slideshow 2

SIRARM Slideshow 3

Academia.Edu

2009-03-25T15:35:00.002+05:30

I have stumbled across my web 2.0 discovery of the month - Academia.edu This is an open access and open editable system which seeks to structure all academic and research institutions across the world.

Users get to choose their institute and department and can then customize their profiles quite thoroughly in order to network with professionals of their same ilk.

This is my profile web page on Academia.edu and it will be constantly updating.

Personally, I think that Academia.edu is a lot like Wikipedia except that Academia compels you to write and edit your own entry. This is actually, quite an innovative idea. The news feed is a good addition and it helps to keep track of events happening within the institute as well as within the research community which you have designated in your research interests.

There are a lot of nifty additions as well. For instance, if someone reaches your home page from any search engine, you get the option to be notified as to how, from where and using which keyword a particular user visited your web page. (You will be surprised at what keywords bring up my home page !!)

As always, with any semantic web tool, the success of Academia.edu will depend largely upon how well it is adapted and used by the target group of users - who traditionally haven't been quite web 2.0 savvy as the rest of the general population.

Invited Talk at Seminar on Spatial Information Retrieval, Analysis, Reasoning and Modelling at the Indian Statistical Institute, Bangalore Centre

2009-03-24T03:56:00.003+05:30

Last week, I had the more fascinating opportunity to interact with some of the eminent scholars and researchers of spatial information science in the Seminar on Spatial Information Retrieval, Analysis, Reasoning and Modelling (SIRARM) held at the Indian Statistical Institute, Bangalore Centre.

As a graduate student of information science in DRTC (a unit of the Indian Statistical Institute under the Computer and Communication Sciences Division) and as a part of my research activities with Prof. BSD Sagar, I was invited to present a talk on the research which we had been doing on UrbanSim.

UrbanSim is a research effort at the Centre for Urban Simulation and Policy Analysis at the University of Washington, Seattle. It is an interdisciplinary effort into the analysis of urban dynamics and urban sprawl using primarily, macro economic data clubbed with models of discrete event simulation.

We have been proud to join this movement powered by the GNU Open Source License. Our research here at ISI primarily focuses on the spatial analysis of appropriate images generated by UrbanSim (for which we shall use tools such as Mathematical Morphology, Fractal Geometry, Rough and Fuzzy Set Theory etc.) Our research is in progress and we are collaborating with researchers from Jadavpur University (Civil Engineering Department)

The slides from my talk can be found here.

Please leave a comment or two if you like (or dislike) this aspect of our research.

Free Academic Goodies

2009-03-05T15:59:00.002+05:30

Whilst web crawling, as I am generally wont to do, I came across 2 startling pieces of the WWW since I was looking for free data sets to extract social networks from and analyze them morphologically.

1. This website links some of the best, freely available data sets available on the internet. Such goodies are god sent gifts for the likes of poor wannabe researchers such as myself as getting access to, and collecting data (under severe financial and time constraints) is nearly impossible. There are dozens of data sets linked on this website, roughly organized by tags for different categories of researchers.

2. This website hosts a XML dump of the entire contents of wikipedia (the mother trove for all prospective information scientists) In fact, the entire meta data of this huge dump has been stored relationally, in the form of tables of PostgreSQL.

All data sets linked to in this post are available freely for academic purposes under the GNU Free Documentation License. I firmly believe that these could prove to be beneficial for prospective undergraduate and graduate students who do not and cannot have access to appropriate data for conducting research.

Paper Submitted to 3rd International AAAI Conference on Weblogs and Social Media

2009-01-24T01:58:00.003+05:30

Research Publications Update:

I submitted an abstract to the 3rd International AAAI Conference on Weblogs and Social Media (ICWSM09). This is a continuation of my research on social network analysis. I have uploaded the abstract here. If anyone can take a look at it and has any ideas or thoughts, please leave your comment here. I would be very pleased to answer to any of your queries.

In my last paper, I spoke about the feasiblity of applying mathematical morphology to study social networks, more specifically, the importance of non network spaces and the geometric variance of the same topologically invariant network. I have extended this study by extracting an appropriate social network from the given dataset by ICWSM09. Specifically, I have used timestamps and blog-comment relationships to draw up an extremely complex social networks. Now, I am in the process of applying the basic morphological transformations of erosion, dilation, opening and closing to a particular subnetwork. There are many different experments which Prof. Sagar and myself are thinking of conducting on this data. We would like to construct some sort of a quantitative prediction model for online social networks since no appropriate models exist right now. Joining us is Dr. Rajan Chattamvelli, our resident data mining expert who is helping us to applying mathematical and statistical data mining techniques to try and refine our extracted social network.

However, as of now, this research is a work in progress and this paper hopes to present some preliminary results of this project to the academic community. Please leave your thoughts on this and any other thing if you so please.

PhD in Business and Technology

2008-12-27T17:02:00.002+05:30

Yesterday, I had a fascinating opportunity to interact with Professor Sridhar Ramaswami, the director of the business PhD program at Iowa State University.

In response to a pre application form which I had filled out in their unique human computer interaction department, I had received a call from Prof. Ramaswami earlier this year. Apparently, Iowa State is starting a new PhD in Business and Technology in its College of Business and is looking to actively recruit new students.

He wanted to meet me since he was coming to Chennai to as a keynote speaker in a conference as well as to teach a short term course in a management school. Yesterday, he came to Bangalore to ISI. We sat in the canteen and spoke for about 3.5 hours.

Upon hearing that I was currently enrolled in a MS program in information science at the Indian Statistical Institute, he encouraged me to consider applying for the Business PhD program for the Fall 2010 session. He informed me about their unique "Management of Information Technology" specialization and the thorough research being done by several professors in my area (social networks, collaborative technologies, virtual communities, enterprise social software etc.)

We spoke at length about the nature of research which I was doing and how I would like to continue it in the future and whether that could result in a good fit between me and the present faculty at Iowa State. He also informed me about the potential benefits of being a business school faculty. We also discussed about life at Ames, academic resources, funding available and the admissions selection procedure. I understood that they would be only taking in 9 people each year initially and that they were only looking for people who brought something unique to the table.

It was a very enjoyable discussion. I also introduced Prof. Ramaswami to my advisor Prof. Sagar and Prof. Naryana of EAU. They spoke for a little bit and Prof. Ramaswami expressed an intention for coming back next summer for some sort of a collaboration between ISI and ISU for doctoral studies.

The ISU business PhD program really intrigues me and I think that it would be a good fit were to apply for PhD in the fall 2010 session.

1st International Conference on Social Network Analysis

2008-12-23T22:22:00.003+05:30

The paper which I sent to the 1st International Conference on Social Network Analysis hosted by the Tata Institute of Social Sciences has been selected for presentation.

This is the invitation letter. I have won a travel grant for this presentation and my registration fees are also going to be waived. I am excited.

My paper is going to be presented on the first day (on Boxing Day) and in much hallowed company. Check the schedule here. In my particular flavor of the subject, I think that it will be very interesting to talk to stalwarts like Prof. Ranjay Gulati of Harvard Business School and Prof. Noshir Contractor of Kellogg School of Management.

My particular paper talks about whether it could be feasible to apply the techniques of mathematical morphology to investigate geometric flexibilities in online social networks. I try to deduce (through a comprehensive literature survey and a case study) whether morphological transformations could be used to determine any sort of a complexity metric for characterizing such networks. The abstract is presented here.

Please leave a comment here if you find this interesting or if you have something to say about it. There will be more on this later.

Data Mining Lecture 5: Genetic Algorithms

2008-11-21T23:34:00.002+05:30

The 5th lecture in the Data Mining lecture series taken by Dr. C. Rajan was primarily based upon the concepts of Genetic Algorithms and their applications and implications upon information science

I would even venture to say that this was the best lecture of the series so far. This is probably because data mining or should I say, more specifically, social mining is an important tool in my area of research and I found genetic algorithms in social mining applications to be quite interesting especially with reference to the quality and shapiness of the data obtained.

The fundamental principles of biological and genetic process which could be applied in algorithmic form in computational science are quite fascinating.

This gives me an opportunity to talk about the exact area where I shall use these data mining concepts (as gleaned from this reading course).

Currently, my research (and hopefully, my thesis area) shall deal in morphological analysis of online social networks. This is highly quantitative in nature as I shall hope to discover or at least derive a standard theory of online social networking behavior. For this, one of the tools shall be data mining whose concepts I shall use to mine social data from relevant data sets. A second variant of this mining could be mining social data straight from the internet and using morphological transformations to derive some sort of an explanation to any significant patterns found in such geo-spatial data in discrete space.

Note to Self: This is just a note to document my thought process as I progress with this research. It would be quite interesting to see how my thought process may or did change with respect to time and more experience.

Top 10 Reasons Why Anyone Would Want To Study in ISI

2008-11-02T22:45:00.003+05:30

This is a very interesting article that came out in Calcutta Times about ISI.

Its a very good and informative read for anyone wishing to study here and for those who wish to know why we claim to be better than some other similar Indian brethren.

To sum up (and at the risk of sounding biased), these are my top 10 reasons why anyone* should want to study in the Indian Statistical Institute:

1. There are no fees (tuition or otherwise) at all, ever.
2. You get a stipend (INR 3000 for undergraduates + INR 4500 for graduates et al.)
3. You have very small class sizes with unity faculty-student ratio (and sometimes even less)
4. You have great extracurricular facilities. (mucho better than some of the other recent institutes which take INR in the lakhs for development facilities which I have seen.)
5. It is under the Ministry of Statistics (and therefore free from all Arjun Singhesque drama)
6. Quota system is almost non existent. (They will only take you if you are good not on the basis of which caste or which religion you are from)
7. The faculty are world class and usually publish in the top journals (Hint: LORs + SOPs)
8. There is 100% placement for the few who need them. (The rest do PhDs and go into academia.)
9. Your fellow classmates are of the order of INMO medallists, God profiles, Prometheus Society members etc. (and therefore great to talk to / work with)
10. Faculty are very approachable and will always work with you in any degree of research as you may see fit.

*Anyone here refers to rational, well defined, distinct individuals or proper Turing machines.

Data Mining Lecture 4 and Logic Programming Lecture 3

2008-10-31T13:06:00.004+05:30

The fourth Data Mining lecture by Dr. Rajan was on Cluster Analysis and it was, by far the hardest of the lot. I found the concepts related to Minkowski distributions and the Mahalanobis distance particularly challenging. However, the applications of cluster analysis to information organization and retrieval were very interesting. Next class, Dr. Rajan is going to discuss genetic algorithms and I need to read up on it to possibly understand anything.

Meanwhile, today, Professor Agostini took a class on logic programming - specifically - Class Logic and its applications. Initially, it was a little difficult to get used to the different forms of logic but later we caught on. It was a very well attended lecture with most of the DRTC faculty attending. I got the opportunity to ask many questions including an intriguing one central to the idea of knowledge representation - which was - Could there be 2 different popositions with the same intensions but different semantic extensions? Turns out, that is the exact practical, research gap which was being worked upon by Prof. Agostini's group at the Knowdive project at the University of Trento.

Many Strands of Thought I:

2008-10-21T15:44:00.002+05:30

This is a conglomeration of different posts which I thought that I would group into one cohesive unit since much has happened in my academic life in the past week that bears chronicling. Perhaps, the following classification is in order:

Data Mining Lecture 3 and the Third Project:

Perhaps, my readers (hopefully, if I have any) are unaware that over the last summer, I did a 3 month internship at “Spinach” - the food retail brand of HDIL (WFR Ltd.) Spinach has already established itself as a major player in western, northern and southern India and its Kolkata setup is its pilot project in the east. I joined Spinach in the first week of May 2008 and participated in a feasibility study and analysis of collaborative management practices (basically, the foundations of building Enterprise 2.0) in its flagship “Spinach Super” store in Kolkata, India. This was the cornerstone of my undergraduate thesis (to be found here) and for which I got the highest grade in the university.

Now, perhaps all of you are wondering what this has got to do with Dr. Rajan’s special lecture series on Data Mining in the Indian Statistical Institute. Now, data (or information or knowledge – call it what you will) mining is one of the fundamental practices of information science and the previous lectures in the series had left me quite interested in finding out the implications or/and applications of data mining techniques to analyze data sets.

During the third lecture which was on decision tree applications, I started to wonder whether I could put any of the stated practices into use. When Dr. Rajan announced that one of the major application areas of data mining was in analyzing retail shopping baskets, I was overjoyed. The reason was that I had exactly such a data set of retail customers of Spinach as reflected by their online shopping behavior. Maybe, I could analyze this information set and gain some meaningful insight into the online buying behavior of customers.

Since Dr. Rajan’s lecture series is only for this semester, I automatically assumed that he would be leaving at the end of this semester. This viewpoint was further strengthened by the fact that Dr. Rajan has just published a book on data mining techniques and applications (since it only costs Rs. 375; with a further discount for students if bought through the ISI library, I am going to buy it and get it autographed by him)

Anyway, so I went to Dr. Rajan’s room and asked him about the best way to analyze retail sets. He asked me to use what he called “association rules” (I wonder what they are) and gave me the references of some open source applications to fiddle around with. So, I told him about the data set and he said that he would be there until August 2009!!

This was an unexpected stroke of luck as then, I can work on this data set throughout the better part of the next semester (with his guidance) and then aim for a half decent journal publication (maybe a decent one if I can co-author with him).

But, first, I have to get the data set from home. (Luckily, I go home at the end of the present month)

Logic Programming Lecture Series and (maybe) the Fourth Project:

This week was a busy one at ISI. We started another special lecture series. This is a set of 10 special lectures spread out over a period of 2.5 months until mid December and is being delivered by Prof. Allessarando Agostini of the Department of Information and Communication Technology at the University of Trento, Italy.

He collaborates with our own Prof. Devika Madalli on the Knowdive project. Prof. Agostini is going to teach us logic programming with special emphasis on contextual logic and OWL (Ontology Web Language.) This is important from the view point of knowledge representation and organization in knowledge based information systems.

His presentation was a very well made one and although we initially struggled a little bit to come to terms with his heavy Italian accent, pretty soon we caught on and were bombarding him with queries. He led us through a history of logic, the different types of logic and a brief introduction to knowledge based systems, ER and UML modeling and its implications to us budding information scientists.

When our turn came to ask questions, we launched a blitzkrieg – not because we wanted to take him down but because his lecture raised a lot of pertinent questions which we need to satiate. This informal discussion continued even well after class timings and finally, the appearance of Prof. Sagar for the next class rescued Prof. Agostini from our clutches.

I am looking forward to the next lecture because my study of the present research at the MIT Media Lab shows that they are working on the Commonsense Project – a huge open access repository of common sense items (viz. “The sun rises in the east.”) and then using this commonsense knowledge as logic premises in expert systems (or by creating different applications of intelligent software agents using this commonsense system)

Maybe in November, I can sit with Prof. Agostini over these series of lectures in order to work out a proposal in using this commonsense and then try and form this into a publishable unit.

The Hyderabad Trip and the DOEACC B Level debacle:

Reliance Computers in Bangalore has some kind of an accreditation problem with DOEACC Society for its B Level (MCA) program. I was running out of time in order to sit for the January 2009 examination. So, I decided to register for the DOEACC program from CAT Academy in Hyderabad.

I have registered for 4 courses in Mathematics (Linear Algebra), Artificial Intelligence and Applications, Principles of Modeling and Simulation and Object Oriented Database Management Systems. Hopefully, I will be able to get good scores on all of them come January if my registration upgrade goes fine.

The better outcome of the entire deal was that I got to spend 2 days in Hyderabad and stayed with my friend Mr. A. Bonnerjee the ICFAI Business School campus.

The Second Project

2008-10-15T23:37:00.002+05:30

Well, the second project sort of bumped into me. My guru, Prof. BSD Sagar is teaching a course in Mathematical Morphology this semester and I, along with Priyanki (and Binit) went to ask him whether he would be interested in taking a second set of special lectures with us where he would just focus on contemporary research being done in the field and possible applications.

Since Prof. Sagar is a close friend of Jean Serra, he indicated that we might want to trace the historical development of the research of Mathematical Morphology in these special lectures by discussing papers written by Serra, as well as other seminal researchers in the field namely Georges Matheron and Pierre Soille.

So, the next day he announces a list of 14 papers which we would take week by week and devour them in order to better understand what was being taught in theory classes. However, the next proposal that Prof. Sagar made caught all of us by surprise. He said that he would like to work with application ideas resulting from these research readings and that, he would try and make them publishable within the year 2009 and named a couple of rather good journals (by impact factor) where we would attempt.

Therefore, I have taken this as my second project. Ideas of pattern recognition, natural language processing and document analysis are swirling in my mind but without a few research readings, I do not think that I would be able to give them any coherent form.

So, I have uploaded the first paper online, here, for everyone to peruse if they so wish and try and help me with any gains gleaned from it. It is the seminal paper by Petros Maragos and Ronald Schafer on transformations in Mathematical Morphology and their usage in binary images.

Paper for 3rd IEEE/ACM International Conference on Information and Communication Technologies and Development 2009

2008-10-05T00:24:00.002+05:30

I should have updated this earlier but I forgot about it. A couple of weeks earlier, I submitted a paper entitled "Collaborative Web Based Construction Informatics - A Learning Sciences Approach in an Indian Context" (a rather unwieldy mouthful of a name I admit) to the 3rd IEEE/ACM International Conference on Information and Communication Technologies and Development.

This conference is going to be held in the Carnegie Mellon University campus in Qatar with Bill Gates as the keynote speaker. As can be established from this as well as the combined IEEE/ACM approach, this conference is big, perhaps the biggest conference that I am attempting to be published in, in my frugal academic career. I hope I get published.

My angle is in construction informatics where I am attempting to prove that web based, collaborative techniques are useful for efficient business management in information conscious developing societies (as opposed to fully developed societies) My paper is basically a case study or a live field experiment (call it what you will) based upon how web based project management can be made more effective in the infrastructure sector in India.

Since I am still a neophyte with respect to big academic events such as this, I have put up my submitted paper online for anybody to review. It is available here. I think that it would be an interesting experiment to see the differences between a double blind peer review in the conference and an open access review here on this blog. If this works, I shall put up all my future paper submissions online for open reviewing.

However, wish me luck on this one.

Information Organization Assignment A

2008-10-04T00:12:00.005+05:30

I had to turn in (or rather, mail in) my first assignment to Prof. Devika Madalli in her Information Organization class.

Her queries to us were to find out what our concepts of classification, library classification and the purpose of classification were and how they had established themselves in our minds.

I had listened to her very attentively in class and had 1. My own class notes and 2. Prof. Madalli's class ppts. However, I didn't want to refer to them to write this assignment as I already had a generalistic view of the topics. What I did, instead was to go to the journal "Knowledge Organization" published by the International Society for Knowledge Organization, supposedly, one of the premier journals in the field and started looking for articles which I could refer to for a philosophical approach to the topic.

Having found a suitable article, I wrote the rest of the assignment in my own words and emailed it to her. I have uploaded it here for reference.

The First Project

2008-10-03T23:18:00.004+05:30

Today, my first ISI project was handed to me by Prof. ARD Prasad. I overheard him asking one of the second years whether they would be interested in some heavy content and context based web site development.

This is because the present DRTC (Documentation Research and Training Centre, a unit of the Indian Statistical Institute) website had not been updated properly in almost 2 years. The website is old and frankly speaking, quite outdated with respect to content and presentation.

Therefore, Dr. Prasad wanted someone with experience to handle this project. So, even though he hadn't asked me personally, I still plodded to his office today in anticipation of whether this project could be handed to me. To my surprise, the Prof. asked me to handle the project completely and independently. I am to use the web application framework, Joomla as a content management system. Initally, I am to use the content of the current website to develop the structure and then, when the knowledge organization of the website is approved, I shall be permitted to experiment with it. (Already, the thoughts of variegated research articles rush through my mind)

I think that this new DRTC website could be a good dataset for playing around with any Web 2.0 or Semantic Web technologies that I have in mind. Perhaps a statistical analysis of website usage? I need to plan it out comprehensively. But first, I have to run Joomla and organize the information in the website.

I have decided to give Dr. Prasad an update every Friday on the project. I am hoping to finish the website and upload it by November. However, this will be a continous improvement process as I shall have to manage the website after creating it. Hopefully, I shall be able to hand in the reins to some other enterprising junior next year.

Data Mining Lecture 2

2008-10-01T21:52:00.003+05:30

Today, was the second in the series of special lectures on Data Mining delivered by Dr. Rajan.

In this class, primarily, we covered the different types of data types and data elements that one may expect when mining data from data warehouses or data marts. Moreover, aspects of metadata - in the sense of the structure of the mined data was also considered.

This class didn't seem as alien as the last one. This was probably due to the fact that we are currently doing a course on "Cataloging and Metadata" and ergo, the concepts of data elements and data storage structures was not completely new.

However, a lot of different, new terminology was learn't - mainly the categorization and types of data types. Essentially, data types are divided into fundamental and extended. Fundamental data types are basically textual or bytestream data. However, extended data can be anything from images, videos, sounds and any combination of them in addition to purely textual data.

A different division of data types into ordinal, numeral, interval and ratio was also discussed in great detail. As opposed to the last class, the audience warmed up to this discussion and questions were asked quite frequently. It also warmed the heart to see the venerable Dr. Raghavan sit down, take notes and ask questions as if this was his first lecture.

I also learn't an important lesson. Lifelong learning is most essential in the lines of research and academics. Even if one isn't in pure academics but is in a more industrial research setting, continous learning is vital for ones success and motivation in the field.

Edit: Removed PPT on Dr. Rajan's request.

Data Mining Lecture 1

2008-09-26T13:34:00.004+05:30

Last Wednesday (September 24, 2008), I attended the first of a series of lectures or special course in Data Mining being conducted by Dr. C Rajan, Visiting Scientist at ISI. Dr. Rajan is an authority on Data Mining with considerable research and teaching experience both in India as well as abroad. This special class will continue for about the next 2 months, (or roughly half a semester).

The first class was an "Introduction to Data Mining" class. It introduced the concepts of data mining, the mathematical and statistical concepts required and the computing tools needed in order to perform Data Mining operations. It also gave us a very broad understanding of the applications of Data Mining.

Whilst many of the math-stat concepts were completely alien to me, the computing concepts weren't, and I came to the conclusion that Data Mining, just as almost all topics in Information Science is a thoroughly interdisciplinary subject. Moreover, I was impressed with Dr. Rajan's presentation which gave a lot of examples of the applications of Data Mining. Some of the applications were self explaintory and logical but many others seemed quite novel and interesting to me.

We do have a separate elective course on Data Mining in our 4th (and final) semester which I am planning to take if additional math-stat electives fail to materialize. I am looking forward to the next lecture to get some deeper insights into Data Mining.

Now, to IKR's final class and then to Brigade Road to meet Trina and Zom for the weekend.

Edit: Removed PPT on Dr. Rajan's request.

The Place

2008-09-24T20:43:00.005+05:30

The weather here, is perennially good. Its always coldish in the morning with a little bit of the mist, quite unlike the heavy smog which I always welcome first thing in the morning back home. In the afternoon, the sun comes out with little or no trace of humidity and the moment it disappears, its cold again, in the evening. The cycle repeats itself every day.

In the morning, I am still sleeping at 9:15 AM when Priyanki wakes me up for the 9:30 AM class. After a bleary eyed first session, I remain in the lab trying to figure out the intricacies of our different courses, and at the beck and call of our faculty. This continues until 4:30 PM when I can finally get a bite to eat, shower, go to the gym and play volleyball until 7:30 PM when I eat tasteless dinner at the mess. Then, the great watch at the computer begins until 2:00 AM at which I time I finally go to bed.

I have no social life and little recreation. What else could I expect in the Indian Statistical Institute?

The Beginning

2008-09-24T17:21:00.005+05:30

Here we go again.

This is yet another beginning. The other blog "Chronicles of Anc" has outlived its purpose. Its purpose, although never explicitly stated was a 3.5 year old vision statement which never successfully fulfilled itself. Therefore, it has not been deleted but merely removed from public vision.

So, here is a new blog which shall be my primary blog from September 2008 onwards. This blog also has a purpose, which although not consistently defined, has a distinct objective that might take a coherent shape as we progress in this web journey.

So, dear Reader. Welcome.