Thanks to CNN!

Sesame Street needs our help! Who is the missing Gordon?

Cain Train stops at Union Station? GOP candidate takes an unexpected stand.

Zuccotti Park is clear. Police force Occupy protesters out.

Check out our first report on the Occupy Baltimore movement in Baltimore’s Inner Harbor. Each week we’ll be featuring different folks who are making the occupation possible and learning about some of the logistical details in order to take our understanding beyond the reasons why people are spending their time with other Occupy participants.

Let us know what you think and be sure to keep an eye out for the next clip in our series.

Ph.D. Dissertation Defense

Identifying and Isolating Text Classification Signals
from Domain and Genre Noise for Sentiment Analysis

Justin Martineau

1:30-4:00 Friday, 18 November 2011, ITE 325b, UMBC

Sentiment analysis is the automatic detection and measurement of sentiment in text segments by machines. This thesis provides methods to identify, characterize, and isolate the sentiment bearing terms to improve textual sentiment classification when there is little or no labeled data for the domain.

We introduce a new theoretical framework that explains the different sources of noise that affect term level sentiment bias. This noise comes from the genre the author communicates in and the domain or general topic that the author is writing about. To understand the affects of domain noise we defined sentimental domain independence and statistically described it in the multi-domain product review data set. This allowed us to design a Domain Independence Verification Algorithm (DIVA) to eliminate this noise and produce a domain-independent sentiment model using data drawn from a variety of different domains. This model is the most accurate method to classify documents in the 25 category product review data set.

Committee:

• Dr. Tim Finin (chair)
• Dr. Marie desJardins
• Dr. Akshay Java
• Dr. James Mayfield
• Dr. Tim Oates

Ph.D. Dissertation Defense

Identifying and Isolating Text Classification Signals
from Domain and Genre Noise for Sentiment Analysis

Justin Martineau

1:30-4:00 Friday, 18 November 2011, ITE 325b, UMBC

Sentiment analysis is the automatic detection and measurement of sentiment in text segments by machines. This thesis provides methods to identify, characterize, and isolate the sentiment bearing terms to improve textual sentiment classification when there is little or no labeled data for the domain.

We introduce a new theoretical framework that explains the different sources of noise that affect term level sentiment bias. This noise comes from the genre the author communicates in and the domain or general topic that the author is writing about. To understand the affects of domain noise we defined sentimental domain independence and statistically described it in the multi-domain product review data set. This allowed us to design a Domain Independence Verification Algorithm (DIVA) to eliminate this noise and produce a domain-independent sentiment model using data drawn from a variety of different domains. This model is the most accurate method to classify documents in the 25 category product review data set.

Committee:

• Dr. Tim Finin (chair)
• Dr. Marie desJardins
• Dr. Akshay Java
• Dr. James Mayfield
• Dr. Tim Oates

Luke 4:14-21 (NIV)

Jesus Rejected at Nazareth

14 Jesus returned to Galilee in the power of the Spirit, and news about him spread through the whole countryside. 15 He was teaching in their synagogues, and everyone praised him.

16 He went to Nazareth, where he had been brought up, and on the Sabbath day he went into the synagogue, as was his custom. He stood up to read, 17 and the scroll of the prophet Isaiah was handed to him. Unrolling it, he found the place where it is written:

18 “The Spirit of the Lord is on me,

because he has anointed me

to proclaim good news to the poor.

He has sent me to proclaim freedom for the prisoners

and recovery of sight for the blind,

to set the oppressed free,

19 to proclaim the year of the Lord’s favor.”

20 Then he rolled up the scroll, gave it back to the attendant and sat down. The eyes of everyone in the synagogue were fastened on him. 21 He began by saying to them, “Today this scripture is fulfilled in your hearing.”

Reflection:

Of all the Old Testament prophets, Isaiah was noted as the one who most vividly predicted the coming of a Messiah. To give a bit of context of what was written on the scroll, we come to a time of political change. Judah finds itself in a time of greatness, being ruled by a king who sought the Lord and obeyed God. But upon his death, the Lord appears to Isaiah in a vision that warned of the judgment to come. But who will go to deliver this message that will certainly be rejected by the people. It is Isaiah who responds, “Here I am! Send Me” (Isaiah 6:8). It is with this reading that we hear Jesus declare his heart and mission - a foreshadowing for what He has come to fulfill - to proclaim “good news to the poor” and “the year of the Lord’s favor” - that which only He could proclaim and do.

Responding to the Transmission:

Jesus came with purpose and mission, to proclaim the good news to the poor and the year of the Lord’s favor. We are living in the time of the Lord’s favor where there is still time to respond to the Good News. But this time will end when Jesus comes back. Let’s recommit our lives today to tell others the Good News that Jesus has come to free us, save us, and heal us. Let’s pray that this would be the crux of our ministry and our walk with the Lord.

After many months of filming on campus, “60 Minutes” featured UMBC on its program this past Sunday. What’s the bad news? It’s all too familiar: the United States continues to graduate too few students, particularly minorities, in science, technology, engineering, and math. The good news: “60 Minutes” focuses on UMBC as a place that gets it right.

We appreciate the recognition – but, more importantly, the opportunity to raise the visibility of this critical issue. View the full segment online here. A transcript is available here.

As you watch, you will see a well-executed production, not the months of taping, research, and editing. You certainly won’t see the messy outtakes. Transformation in higher education happens in much the same way. It’s a process, not a product. What started in 1988 with an idea for a select program for African-American males in science, math, and engineering – the Meyerhoff Scholars Program – has blossomed into a broad culture of student achievement at UMBC.

Lessons from the Meyerhoff program, for example, inspired scholars programs for undergraduates in the arts, humanities, public affairs, and teacher education. More recently, the Meyerhoff experience with group study informed course redesigns for first-year courses in science, engineering, and psychology. And, in turn, lessons from that initiative now have UMBC working to infuse active learning throughout its curriculum.

But these few examples aren’t the story. What we have worked toward over the past 25 years at UMBC is a culture that goes beyond “project mode” when it comes to improving undergraduate education.  More than a particular program, department, or initiative – we are an entire community that works relentlessly to improve the educational experience we provide.

Our faculty, staff, students, alumni, and partners have worked tirelessly, and often behind the scenes, to breathe life into that work. UMBC’s successes are your successes. Thank you for your continued support, and enjoy the spotlight!